Culture and Reproduction: An Anthropological Critique of Demographic Transition Theory 9780429038693

Citation preview

Culture and Reproduction

ERRATA

for Culture and Reproduction: An Anthropological Critique of Demographic Transition Theory by W. Penn Handwerker The following lines were omitted from the text. They connect pages 200 and 201, concluding the last sentence on page 200 and beginning the first line on page 201. as the armed forces drained supplies from the domestic economy. 3 In this way, through the late 18th and early 19th centuries, the London meat market gradually extended itself to the most remote corners of the British Isles (Orwin and Whethan 1971:25; Trow-Smith 1959; Ross 1982). As English industrialization advanced, moreover, the need to maintain food supplies was heightened by the frequency with which scarcity and high prices became a focus for working-class activism. Irish (and Scottish) foodstuffs-later supplemented by meat and grain from the Americas and Australasia-moderated prices, thus restraining that activism and the demand for higher wages. Because these imports held down the costs of manufacturing, England maintained a competitive edge in the international market (cf. Ainsworth-Davis 1924:81). The elaboration of Ireland's export

About the Book and Editor World population growth, especially its fertility component, poses a major dilemma for policymakers throughout the world. However, theoretical developments in demography have not yet provided a solid foundation for understanding contemporary population processes. From an anthropological perspective, the current micro-level models do not properly recognize the cultural and biological constraints within which people make reproductive decisions. On the macro level, demographic transition continues to be linked to processes of"modernization." Arguing that it is necessary to readdress micro-level issues in light of the culturalhistorical variability of particular places and times and to explore linkages between macro-: and micro-level phenomena through which population processes work themselves out, the contributors point the way to new theoretical formulations of the concept of culture, the nature of macro/micro linkages, and methods of placing demographic theory within the more encompassing framework of evolutionary theory. W. Penn Handwerker is professor of anthropology at Humboldt State University.

Culture and Reproduction An Anthropological Critique of Demographic Transition Theory edited by W. Penn Handwerker

I~ ~~o~~!~n~~~up LONDON AND NEW YORK

First published 1986 by Westview Press, Inc.

Published 2018 by Routledge 52 Vanderbilt Avenue, New York, NY 10017 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Routledge is an imprint of the Taylor & Francis Group, an informa business

Copyright © 1986 Taylor & Francis

All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe.

Library of Congress Cataloging in Publication Data Culture and reproduction. Includes bibliographical references. I. Demographic transition. 2. Fertility, Human. 3. Demographic anthropology. I. Handwerker, W. Penn. HB887.C85 1986 304.6'32 86-5671 ISBN 13: 978-0-367-00883-3 (hbk)

Contents Foreword, Dudley Kirk . .................................... xi Preface................................................... xiii Notes on Contributors .................................... xv Culture and Reproduction: Exploring Micro/Macro Linkages, W. Penn Handwerker. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 PART ONE MICRO-LEVEL ISSUES 2

Rationality and Models of Reproductive Decision-Making,

Paul V. Crosbie........................................... 30 3

Biological Determinants of Fecundity and Fecundability: An Application of Bongaarts' Model to Forager Fertility,

Edwin N. Wilmsen ...........................·............ 59 A Comment on the Determinants of !Kung Fertility, John Bongaarts ........................................... 83 Rejoinder to Bongaarts, Edwin N. Wilmsen ....................................... 87 4

"Natural Fertility" as a Balance of Choice and Behavioral Effect: Policy Implications for Liberian Farm Households, W. Penn Handwerker . .................................... 90

5

Transformation in the Natural Fertility Regime of Western Alaskan Eskimo,

Jean M. Brainard and Theresa Overfield . . . . . . . . . . . . . . . . . . 112 vii

6

Price or Production? Domestic Economies, Household Structure, and Fertility in a Guatemalan Village,

Mary E. Odell .......................................... 125 7

Fertility and Historical Variation in Economic Strategy Among Migrants to the Lacandon Forest, Mexico,

Debra A. Schumann ..................................... 144 8

Cultural Constraints on Fertility Transition in Tunisia: A Case-Analysis from the City of Ksar-Hellal,

Liesa Stamm and Amy Ong Tsui. ........................ 159

PART TWO MICRO/MACRO-LEVEL LINKAGES 9

Resources, Rivalry, and Reproduction: The Influence of Basic Resource Characteristics on Reproductive Behavior,

Brian Hayden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176 10

Potatoes, Population, and the Irish Famine: The Political Economy of Demographic Change,

Eric B. Ross . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196 11

Contraception, Ideology, and Policy Formation: CohQrt Change in Dublin, Ireland,

Kevin R. O'Reilly. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

2~1

Economic Diversity, Family Strategy, and Fertility in a Mexican-American Community,

Linda M. Whiteford. ..................................... 237 13

Labor Expropriation and Fertility: Population Growth in Nineteenth Century Java,

Paul Alexander . ......................................... 249 14

The Political Economy of Fertility Regulation: The Kusasi of Savanna West Africa (Ghana),

David A. Cleveland ...................................... 263 viii

15

Agricultural Intensification and Fertility in The Gambia (West Africa),

Peter Wei/ .............................................. 294

16

Temne Fertility: Rural Continuity, Urban Change, Rural-Urban Differences, and Public Policy Problems,

Vernon R. Dorjahn ...................................... 321

References Cited ............................................. 350 Index ..................................................... ... 383

ix

FOREWORD The world population problem is much too important and complex to be left to a single discipline of study. This book is a vigorous attempt to bring together two very different ways of looking at the problem into a single, unified approach. The two disciplines are demography and anthropology. One is tempted to see a "natural" antipathy between the two with reference to such matters as human reproduction. Demography is a highly quantified discipline; anthropology traditionally is one that deals with qualitative, often highly subjective matters. Demography deals with "hard" data, anthropology with "soft." Demography focuses on the objective measurement of behavior, the "how," anthropology on the "why." Demography seeks to isolate cause and effect in empirical generalizations or "laws"; anthropology views its subject matter more in terms of the complex interaction of variables much more difficult to enshrine in cross-country generalizations. Demography is essentially one dimensional in time, picturing all countries as on a single continuum of change from traditional to modern and especially to Western demographic behavior. Anthropology looks at reproductive and other behavior in depth with a stereoptic vision. Demography looks for the common aspects of reproductive behaviors in different societies; anthropology often seems to be more interested in the different, the unusual and the exotic rather than in quantifiable central tendencies. Partisans of both disciplines will doubtless find these characterizations to be simplistic, but hopefully at least indicative of the apparent gulf between them. This volume is an able effort to bridge the gap and bring together the two very different approaches toward a unified theory that combines the best each has to offer. The editor and author do not assert that this has been fully achieved in this book, but that hopefully it is a significant step in the right direction. The particular focus of this book is the theory of the demographic transition, in which modern population change is viewed as moving along a continuum from the high birth and death rates of most traditional societies to the low birth and death rates now found in all industrialized (i.e., more developed) countries. To document the position of different countries along this continuum there have been impressive field studies such as those of the World Fertility Survey which were conducted on a comparable basis in

xi

some 25 countries. These are large national sample surveys analyzed by demographers in highly sophisticated quantitative terms, though in fact they deal with behavior derived from some of the most intimate attitudes and motivations which may be very different for different individuals in different societies. There is little question about the fact of the demographic transition in very broad terms. But, as earlier suggested, demographers have chiefly dealt with the "how" rather than the "why." Thus demographers point with a certain justifiable pride to the sophisticated and meticulous measurement of levels and trends in fertility and mortality and to multivariate analysis of the association between socio-economic measures, especially education, with progress in the reduction of fertility. Demographers are much more involved with the objective mechanics than with the subtleties of attitudes, motivations and behaviors of individuals that collectively bring this about. They all too often ignore the cultural and historical content which promotes or retards reduction of the birth rate. For example, the use or non-use of contraception and abortion for birth control is obviously sensitively affected by the cultural context. Why fertility occurs at all and why it occurs at different rates and in different ways in different countries is inadequately explained by demographers. Here is the proper domain of anthropologists; they are better trained and experienced in comprehending the complex cultural and historical interactions that determine changing human motivations and behavior relating to reproduction. Here, too, is a basis for bringing together the perspectives of anthropology and demography. The microdemography of anthropology places reproductive behavior within its particular culturehistorical context and helps us see the processes that create patterns in the aggregated data on which demographers focus. Examining relationships between micro-level processes and the macro-level constraints that may shape those processes may help us better understand both why and how demographic transition comes about. The essays that comprise this book report on a wide diversity of cultures and specific problems. Some essays focus on the rnicrodemographic processes of specific times and places; others examine linkages between microlevel processes and macro-level constraints. The editor, Professor Handwerker, has done a painstaking job in editing, organizing and transforming a rich but highly diverse set of contributions into a very useful whole. However, each contribution stands on its own; a specialist in any area or problem can readily identify the works most relevant to his or her interests.

Dudley Kirk Xll

PREFACE This book originated in a conference on Culture and Reproduction held 2 December 1981 at the University of California, Los Angeles. The purpose of the conference was to bring together anthropologists and demographers to discuss exciting conceptual changes in demographic theory. The demographers Louis Henry, John Bongaarts, J.C. Caldwell, and Richard Easterlin had made considerable progress toward a major reconceptualization of demographic transition theory, particularly with regard to understanding the major patterns of human fertility and circumstances under which fertility patterns changed. Their work appeared to lay the groundwork for a comprehensive theory of fertility and fertility transition. Many of the questions their speculations raised, however, appeared to be best answered by cross-cultural work. With the exception of Caldwell, few demographers appeared to be aware of the data and perspectives of anthropologists. The effect was to leave glaring conceptual lacunae in otherwise promising models. The anthropologists brought data and perspectives on human cultural and biological variation across space and time, and the demographers brought expertise in demographic methodologies and theory. The outcome was an exceptionally stimulating exchange of views. The essays in this volume reflect that exchange and attempt to contribute to the intellectual ferment that now characterizes the study of human fertility by documenting cross-cultural variation in the patterns and determinants of fertility change. Most of the papers originally presented at the conference appear in this volume. Both to address issues we were not able to consider at the conference, and to broaden the geographical scope of our presentation, several papers have been added to the ones originally presented. The content of all papers reflects the growth in ideas that has occurred over the past few years. The introductory chapter is especially indebted to the authors contributing to this volume. Everyone associated with the original conference contributed to its success. The demographers John Bongaarts, Bill Butz, Jack Caldwell, and Dudley Kirk, took time from busy schedules to contribute in very positive and challenging ways. At UCLA, Susan Scrimshaw and her secretary Julia Urla greatly facilitated making technical arrangements for the conference. Xlll

The Humboldt State University Foundation supported the costs of the conference not borne by registration fees. I am especially grateful for the personal assistance and encouragement offered by Jim Hamby, the Humboldt State University Foundation Manager. Ellen Agard, then an undergraduate at HSU, helped with some planning details in Arcata, and supervised the on-site registration in Los Angeles. Judy Hampton and Bea Hulsebus, the secretarial staff for the Department of Sociology, Anthropology, and Social Work at Humboldt State University, put up with my editing activities with exceptional grace. They, and John Mulvaney, the managing editor of the Humboldt Journal of Social Relations, were particularly helpful in allowing me to monopolize for nearly a year what was then our only word processor. And they saved me considerable time by preparing the References Cited section and in adding papers and miscellaneous odds and ends to the document file. Prior to their inclusion in this volume, all papers have undergone a peer review. The finished volume has likewise. I thank colleagues who took valuable time to produce thoughtful, useful reviews of papers included in this volume: Paul Alexander, Tony Carter, George Cowgill, Don Dumond, Rose Frisch, Mary Granica, Joel Gregory, Gene Hammel, Sarah Harbison, Sandra Huffman, Barry Isaac, Young J. Kim, Ann Millard, Ted Ruprecht, Jane Schneider, and two anonymous referees. The papers are uniformly better for their efforts. I also thank the authors of this volume, for paying some attention to deadlines even when it proved impossible to meet them, but especially for their patience with what at times must have seemed the substantial editing liberties I took with their manuscripts. W. Penn Handwerker Trinidad, CA

xiv

NOTES ON CONTRIBUTORS Paul Alexander is a New Zealander who teaches Social Anthropology at the University of Sydney, NSW 2006, Australia. His 1973 Ph.D. from the Australian National University was a study of a fishing village in Sri Lanka. He has published SRI LANKAN FISHERMEN: RURAL CAPITALISM AND PEASANT SOCIETY (Australian National University Press, 1982), and several papers on the political and economic anthropology of Sri Lanka. For the past five years, his research has centered on the political economy of colonial and post-colonia Java. John Bongaarts is a Senior Associate at the Center for Policy Studies, The Population Council. His 1972 Ph.D. from the University of Illinois was in the area of physiological and biomedical engineering. He has written extensively on sociobiological determinants of fertility, family demography, and mathematical demography. He co-authored (with R.G. Potter) FERTILITY, BIOLOGY, AND BEHAVIOR (Academic Press, 1982) and has major papers in THE DETERMINANTS OF FERTILITY IN DEVELOPING COUNTRIES (R.A. Bulatao and R.D. Lee, eds. Academic Press, 1983). Jean Brainard is Assistant Professor in the Department of Anthropology at The Ohio State University. Her 1981 Ph.D. in biological anthropology from the Binghamton campus of the State University of New York was a study of reproduction among Turkana pastoralists in eastern Africa. Jean has published several papers on her African research. She is currently planning a comparative study of maternal and child health and nutrition encompassing the principal ethnic components of the United States, including southeast Asian refugees. David Cleveland is Co-Director of The Center for People, Food and Environment in Tucson, Arizona, an organization devoted to research and action on sustainable food systems. His 1980 Ph.D. from the University of Arizona was a study of agricultural and ecosystem change among the Kusasi of northeastern Ghana. A specialist in drylands ecology, David has published on drylands production in Arizona as well as West Africa and is co-authoring a book on household gardens in drylands. He is currently conducting field research on household food production in Egypt. Paul Crosbie is Professor of Sociology at Humboldt State University on California's North Coast. His 1969 Ph.D. in sociology from Stanford University was a study of power and decision-making. He has conducted research on social interaction and exchange, power and dominance, and fertility decision-making and contraceptive behavior. He has published extensively in the area of small groups theory and methodologies and has edited a major text in the field: INTERACTION IN SMALL GROUPS (Macmillan, 1975). Currently, he is conducting research that seeks to understand why some people actively protected Jews during the Holocaust, and why others did not. XV

Vernon Dorjahn is Professor of Anthropology at the University of Oregon. His 1954 Ph.D. in ethnology from Northwestern University was one of the earliest major demographic studies by an anthropologist, a survey of the demographic aspects of polygyny in Africa. He has conducted research on a wide range of topics in ethnology and demography, including history, political organization, polygyny, migration, mortality, and fertility, principally among the Temne of Sierra Leone, and has published widely. His current research focuses on 19th century social organization, polygyny, and divorce in Sierra Leone. Penn Handwerker is Professor of Anthropology at Humboldt State University on California's North Coast. His 1971 Ph.D. from the University of Oregon was a study of food production and marketing in the Republic of Liberia, West Africa. His principal theoretical and research interests center on human population ecology, evolution, and the political economy of development. He has published widely on the topics of food distribution, food production, and fertility, based largely on West African research, and has published a paper reviewing prehistoric demographic trends. More recently, Penn has undertaken research on agricultural sector management and corruption in West Africa. Currently, he is undertaking a field project in the Eastern Caribbean to test a theory of fertility transition. Brian Hayden is Professor of Archaeology at Simon Fraser University. His 1976 Ph.D. in archaeology from the University of Toronto was a study of lithic technology among contemporary Aboriginal populations in Australia. He has a long standing interest in the dynamics of population regulation, especially the cultural controls, and he has conducted ethnoarchaeological research in Australia and Guatemala. Brian has published major papers on the topics of prehistoric population control, the concept of carrying capacity, huntergatherer organization, and technological evolution, has edited LITHIC USE WEAR ANALYSIS (Academic Press, 1979), and has authored PALEOLITHIC REFLECTIONS (Australian Institute of Aboriginal Studies, 1979) and (with Aubrey Cannon) THE STRUCTURE OF MATERIAL SYSTEMS (Society for American Archaeology, 1979). His current research interests focus on cultural ecology and the emergence of new technologies during the Paleolithic, as well as changes in cultural complexity. Mary Odell is a Senior Research Associate at BOKONON Systems in Washington, D.C. Her 1978 Ph.D. in biological anthropology from Temple University focused on the effects of agricultural innovations on human fertility in Guatemala. She has published several papers based on her Guatemalan research. More recently, she has developed research on factors affecting the adoption of fertility control in Northeast Brazil. Currently, Mary is conducting research on public policy aspects of welfare and child support enforcement in the United States. Kevin O'Reilly is a Research Anthropologist at the Centers for Disease Control, where he has worked in the Division of Reproductive Health, and currently in XVI

the Division of Sexually Transmitted Diseases. His 1980 Ph.D. in medical anthropology from the University of Connecticut was a study of family planning in Dublin, Ireland. More recently, Kevin has conducted research on traditional midwives as promoters and distributors of contraception in Kenya. Kevin is a past chair of the Council for Anthropology and Reproduction. His current research focuses on AIDS and sexually transmitted diseases in the United States. Theresa Overfield is Professor and Director of Research at the College of Nursing, Brigham Young University. Her 1975 Ph.D. in physical anthropology from the University of Colorado extended prior training in nursing and public health. She has conducted research among Alaskan Eskimo and has published extensively in the area of medicine and biological variation. Her book BIOLOGICAL VARIATION IN HEALTH AND ILLNESS (Addison-Wesley) appeared in 1985. Currently, her interests are divided between administration and research in medical anthropology. Eric Ross is Adjunct Assistant Professor of Anthropology at the University of Florida but resides in London. His 1976 Ph.D. in ethnology from Columbia University was a study of the Achuara-Jivaro Indians in Peru. He has published several papers and has edited BEYOND THE MYTHS OF CULTURE: ESSAYS ON CULTURAL MATERIALISM (Academic Press, 1980). He has two forthcoming books, FOOD AND EVOLUTION (co-edited with Marvin Harris, Temple University Press), and DEATH, SEX, AND FERTILITY (coauthored with Marvin Harris, Columbia University Press). Eric's most recent research concerns the evolution of the British capitalist economy and its role in the underdevelopment of Scotland and Ireland, with special attention to its dietary and demographic effects. A book reporting on this work, ENGLISH FEAST, IRISH FAMINE: THE POLITICAL ECOLOGY OF AGRICULTURAL CRISIS, is currently in progress. Debra Schumann is a Postdoctoral Fellow in the Department of Population Dynamics, The Johns Hopkins University. Her 1982 Ph.D. in ethnology from Southern Methodist University was a study of fertility and agricultural change in southern Mexico. She has served as a consultant for UNDP, the Israeli Ministry of Health, the U.S. Peace Corps, Indiana University, and Planned Parenthood. She has conducted research on the economic impact of tropical diseases, drug abuse, agriculture, and the utilization of health care services. Debra has published several papers and has co-edited (with William Partridge) THE HUMAN ECOLOGY OF TROPICAL LAND SETTLEMENT: LATIN AMERICAN PERSPECTIVES. Her current research plans focus on the epidemiology of child mortality in Nigeria. Liesa Stamm is Assistant Director of Program Planning and Review in the Department of }jigher Education for the State of Connecticut. Her 1980 Ph.D. in ethnology from the University of Illinois was a study of social and cultural change among Tunisian women. She has conducted research both in Tunisia

xvii

and the United States on a wide range of topics: women's social power and influence, perinatal practices and fertility control behavior, and the service needs of the homeless. Liesa has published several papers and has co-edited (with Carol Ryft) SOCIAL POWER AND INFLUENCE OF WOMEN (Westview, 1984). Her most recent research focuses on the effectiveness of intervention services in the prevention of child abuse.

Amy Ong Tsui is Assistant Professor, Maternal and Child Health Department,

School of Public Health, and Fellow, Carolina Population Center, University of North Carolina. Her 1977 Ph.D. in sociology from the University of Chicago was a study of family formation processes among United States marriage cohorts. She has published extensively on issues in fertility and family planning in the Third World and on family and household formation patterns in the United States. Her current research is on consequences of fertility regulation in Egypt, maternal and child health behaviors in Zaire, and estimation of family planning needs for Latin American countries.

Peter Well is Associate Professor of Anthropology at the University of Delaware. His 1968 Ph.D. in ethnology from the University of Oregon was a study of the political system of The Gambia. He has conducted research on topics focused on 19th and 20th century socioeconomic change in The Gambia, with particular emphasis on changes in agricultural production technologies and the organization of production. A secondary area concerns political communication processes. Recently, he has conducted extensive applied research focused on the impact of agricultural development initiatives planned by public agencies on the socioeconomic organization and cultures of peoples intended to benefit from them. Peter has published extensively on all aspects of this work and has coedited a book on rural development. Unda Whiteford is Associate Professor of Anthropology, and Leader of the Medical Track for the Applied Anthropology Program, at the University of South Florida. Her 1979 Ph.D. in medical anthropology from the University of Wisconsin-Milwaukee was a study of family planning and reproductive strategies in south Texas. She has conducted research focused on health care delivery systems, family planning, and reproductive strategies in England, Denmark, Mexico, and,the United States. Linda has published in the areas of community health evaluation, health policy development, and ethics and reproduction. She is a past chair of the Council for Anthropology and Reproduction. Her current research interest is in primary health care and the development of health care policy in the Dominican Republic.

Edwin WUmsen is Professor of Anthropology at the African Studies Center at

Boston University. His 1967 Ph.D. from the University of Arizona developed a methodology for increasing the value of material items, particularly prehistoric lithics, in the study of cultures. His research interests have centered around the relations of economy and society and in the reproduction of resultant social forms; his investigations of the connections between economic production,

xviii

nutrition, and fertility stem from those interests. He has conducted extensive field research in southern Africa and has published widely. His current research continues to pursue these themes.

xix

Culture and Reproduction

1

Culture and Reproduction: Exploring Micro I Macro Linkages W. Penn Handwerker

Introduction The essays in this volume describe micro-level dimensions of human reproductive patterns and explore linkages between micro- and macro-level phenomena. The first part of this book focuses on micro-level issues. Crosbie takes a fresh look at rationality models of reproductive decisionmaking and their critics and tests hypotheses offered by both. He finds that rationality models predict demand for children, but not reproductive behavior. Wilmsen addresses the controversy surrounding the biological determinants of fertility by arguing that the conventional formulation pitting Frisch's critical-fat hypothesis against lactation-induced amenorrhea mis-specifies the problem. Handwerker models the determinants of fertility for a ''natural fertility'' population in West Africa. Brainard and Overfield trace changes in Western Alaskan Eskimo fertility between 1945 through 1974 and find that the adoption of family planning services has not been accompanied by fertility transition. Odell tests competing hypotheses offered by Easterlin and Caldwell with data from Guatemala. Schumann's case-study from Chiapas finds that historical and cross-sectional variation in economic strategies has not determined fertility; rather, variations in fertility have determined which families could take advantage of economic opportunities requiring increased family labor inputs. Stamm and Tsui link Tunisia's incipient fertility transition to a variety of cultural factors and suggest that a completed fertility transition presupposes that family size becomes dissociated from evaluations of adult prestige. 1

2

HANDWERKER

The second part of this book explores linkages between micro-level processes and the macro-level constraints that shape those processes. Hayden provides an overview of relationships between human population growth and resource availability over the last 20,000 years that links changes in fertility to competition generated by increasing availability of resources. Ross re-evaluates the Irish case on which the Malthusian model has depended so strongly for support and argues that Irish fertility, the potato, and The Great Famine of 1845-49 owe their origin to a common determinant: the policies of (largely) absentee landlords responding to the vicissitudes of the English market. O'Reilly takes up where Ross's paper leaves off, outlines the recent history of family planning legislation in Ireland, and documents a cohort change in fertility among women in Dublin stemming from the public controversy surrounding family planning legislation. Whiteford analyzes data on a Mexican-American community in southern Texas with an eye to understanding not merely why MexicanAmerican fertility is so high relative to other components of the U.S. population, but also why and how fertility varies within this community. She argues that low fertility reflects the ability to take advantage of educational and economic opportunities. Alexander re-evaluates the rise of the Javanese population after 1830, arguing that population growth is attributable primarily to increases in fertility that stemmed not from intentional responses to an increased demand for labor but to a reduction in the. duration of breastfeeding that accompanied changes in women's labor activities. Cleveland and Weil present case-studies of the relationships. between the intensification of agriculture and increases in fertility in West Africa. Both document changes in family organization and intentional increases in fertility predicated, at least in part, on declining infant mortality. Dorjahn analyzes changes in rural and urban fertility, and rural-urban fertility differences, among the Temne of Sierra Leone. He raises the question of the potentially deleterious impact of continued high fertility and relates this concern to discontinuities among three, potentially very different, perspectives: (1) the wishes of a central government, (2) the wishes of the people with whom a fieldworker has lived and worked, and (3) what the fieldworker might perceive to be social science truth. These essays thus critique two major sets of standard demographic theory-demographic transition theory and the decision models of reproductive behavior that have come to underpin the modernization interpretation of fertility transition. This introductory chapter reviews contemporary demographic transition theory, summarizes these critiques to identify some of the more serious deficiencies of standard theory, and explores the possibility of specifying linkages between the macro-level determinants of fertility and the micro-level processes through which fertility transition

CULTURE AND REPRODUCTION

3

works itself out. The theme that unites these otherwise diverse essays is that fertility transition in the contemporary world comes about when personal material wellbeing is determined less by personal relationships than by formal education and skill training. This transformation occurs when changes in opportunity structure and the labor market increasingly reward educationally-acquired skills and perspectives, for these changes have the effect of sharply limiting or eliminating the expected intergenerational income flows both from children, and from the social relationships created by or through the use of children. The adverse effect of this reversal of the intergenerational wealth flow on parental material well-being provides the incentive for the sharp limitation of family size that has been the singular feature of the modern demographic transition. This hypothesis fills a gap in wealth flows theory (Caldwel11982), suggests solutions to a number of puzzles raised by conventional demographic theory, and permits us to account for historical, regional, and social class differences in the onset and pace of fertility transition. This hypothesis also implies that we have been attempting to explain demographic transition with strategies that by their very nature cannot provide the answers. These answers may come when we place demographic theory within the more encompassing framework of evolutionary theory.

The Modernization Interpretation of Demographic Transition The extraordinary growth rate of the human population in recent years (about 1.9 percent annually between 1950 and 1980) has its origins some 200 years ago. About 1750 the populations of Western Europe which had for centuries experienced high levels of mortality and fertility began to experience a decline in mortality, and after 1850, a decline in fertility. An exceptional mortality sequence from France (Bourgeois-Pichat 1965), for instance, documents a decline in infant mortality from about 250 in the 1700s to around 170-180 between 1805 through the late 1800s, and a subsequent fall to around SO in 1950. European marital fertility declined from levels normally ranging from about 7-10 to levels in the neighborhood of 2-4, and total fertility rates close to, and in some cases below, replacement levels. This demographic transition from high to low birth and death rates occurred together with (1) the transformation of European society from an agrarian to an industrial regime, and (2) a transformation from relatively complex extended family units which Shorter (1975) has characterized as mechanisms for the transmission of property and position, to smaller, less differentiated units based on the emotional attachment of husband and wife, and within which children moved from a position of marginality to

4

HANDWERKER

centrality. These and related social, religious, political, and economic transformations were the focus of the work of the early social theorists: Comte, Durkheim, Weber, and Marx. The question of the impact of the industrial system on the structure and nature of human relationships remains central to social enquiry generally, and to demographic transition theory specifically. Today, as in the 1970s when major critiques appeared (e.g., Polgar 1972, Giddens 1976, Porteres 1976, Caldwell 1976, and prior to these notably Frank 1967), the prevailing social and demographic paradigm is still based on the distinction between Traditional and Modern Societies elucidated by the classical social theorists, and the problem is still taken to be understanding the nature of the two and of the transition between them. It continues to be widely accepted that fertility transition should follow from "development"-usually defined in terms of industrialization and Westernization/Modernization-and a transition from a Traditional to a Modern society (e.g., Tabah 1980). However, largely on the basis of work by Louis Henry (1961; cf. Knodel1983, Leridon and Menken 1979), John Bongaarts (1982, .1983), and Richard Easterlin (1978, 1983; cf. Bulatao 1983), interpretation of this paradigm-for instance, as developed in the recent volumes on The Determinants of Fertility in Developing Countries (Bulatao and Lee 1983)-has been updated. It has become possible to redefine the nature of the social types used to model fertility transition, to specify the niodels more realistically than by using, e.g., Parsons' pattern variables and the full set of Davis-Blake (1956) intermediate fertility variables, and to change the propositions to be examined. High fertility, for instance, is no longer viewed as "uncontrolled" and "irrational/' supported merely by religious proscription, curious taboos, and unthinking acquiescence to cultural norms. High fertility is thought of instead as stemming from reproductive patterns in which the childbearing period is not truncated-i.e., Henry's "natural" fertility regime. In such regimes, the age pattern of childbearing is determined primarily by the age- specific proportion of sterility and the intrauterine mortality rate, which may vary insignificantly across populations. After age 30, age-specific fertility declines until menopause at average age 48-50, but does so relatively slowly. Consequently, age-specific fertility schedules display a characteristic concave shape and the ratio of the observed age-specific fertility schedules to the standard natural fertility schedule constructed by Coale and Trussell (1971, 1975) should be relatively constant by age, generally approximating unity. Bongaarts' work suggests that total fertility in such regimes is determined primarily by only two intermediate fertility variables: the proportion of women married and the average duration of postpartum infecundability. Postpartum infecundability itself is largely a function of the average

CULTURE AND REPRODUCTION

5

duration of breastfeeding or, in some societies, the average duration of postpartum abstinence. Easterlin argues that the parameters of these intermediate fertility variables, and the number of children who survive to adulthood, respond to income and costs and will be a function of a couple's demand for children, the supply of children, and the application of techniques to regulate supply. Natural fertility regimes are defmed by the presence of excess demand where the number of children one wants to survive to adulthood is greater than the maximum number of children one can actually raise to adulthood. Hence, techniques to regulate supply are not important determinants of supply, the total fertility rate will be the maximum number of conceptions which can be brought to full term, and the number of children who actually reach adulthood is the maximum number one can raise to adulthood. Parental utility in natural fertility regimes is a simple function of the goods and services they consume and the number of children who actually survive to adulthood. The supply of children is a function of the goods and services consumed. The couple's potential lifetime income is equal to the market costs for the goods and services a couple consumes, plus the costs of raising to adulthood the children who survive to that point. Fluctuations in the level of natural fertility, and fertility differentials within particular regimes, thus are modeled as functions of wealth and costs. Conversely, low fertility is no longer contrasted with pre-transition fertility as being rationally controlled and reflecting an orientation toward achievement and universalistic norms. Low fertility is thought of instead as stemming from reproductive patterns in which the childbearing period is truncated-i.e., Henry's "target" fertility regime. Fertility transition is defined by reference to the transition from "natural" to "target" fertility as women apply techniques to regulate supply and to truncate their childbearing period at increasingly earlier ages. The age pattern of childbearing changes as the application rate and effectiveness of contraception and abortion increase. Age-specific fertility falls increasingly rapidly after age 30. Consequently, the age-specific fertility schedule of "target" fertility populations displays a characteristic scallop at its tail and the ratio of the observed age-specific fertility schedule to the Coale-Trussell standard for natural fertility populations should diverge sharply below unity. Thus, total fertility becomes a function of four intermediate fertility variables: the proportion of married women, the average duration of postpartum infecundability, the rate and effectiveness of contraception, and the rate of induced abortion. As in natural fertility regimes, both the intermediate fertility variables and the number of children who survive to adulthood respond to income and costs and will be a function of a couple's demand for children, the

6

HANDWERKER

supply of children, and the application of techniques to regulate supply. Target fertility regimes are defined by the presence of an excess supply of children where the number of children one wants to survive to adulthood is smaller than the maximum number of children one can actually raise to adulthood. Hence, techniques to regulate supply are important determinants of supply. Parental utility becomes a function of, in addition to the goods and services consumed and the number of children surviving to adulthood, the psychic disutility attached to using supply regulating technologies. A couple's budget constraint now includes the costs of obtaining information about .regulation technologies plus the market costs incurred while averting a given number of births or decreasing the number of children surviving to adulthood. Decisions to avert births reflect the relationship between the goods and services one wishes to consume, the number of children one wishes to raise to adulthood, the psychic disutility of actually using regulating technologies, and the actual costs of regulating the supply of children. The difference between the maximum number of children that could survive to adulthood and the number of children actually wanted constitutes a measure of the incentive to apply regulating technologies. The determinants of the change from an excess demand to an excess supply are to be sought in such structural factors as improvements in public health care, growth in formal education and the mass media, urbanization, the availability of new goods and services, and growth in per capita income (Easterlin 1978: 110; cf. Notestein 1945, Davis 1963, Heer 1983, Easterlin 1983).

Current Puzzles Increasing conceptual sophistication has not, however, resolved some of the most fundamental issues surrounding the demographic transition. Recent work in Europe (e.g., Coale 1975, Knodel and van de Walle 1979), for instance, calls into question the connections between technological, familial, and demographic change that have been the basis for interpreting fertility transition. With only minor exceptions, fertility transition began in Western Europe within a period of 30 years between 1880 to 1910 under a diverse set of socioeconomic and demographic indices (percent literate, per~ cent rural, percent in cities over 20,000 population, percent of male labor force in agriculture, infant mortality). Findings presented in this volume confirm this diversity by revealing that most indices of "development" or "modernization" such as improvements in public health care, the availability of new goods and services, and growth in per capita income, are likewise factors whose putative linkages with fertility are either spurious or

CULTURE AND REPRODUCTION

7

inconsistent. Innovations in public health and medical care that might be presumed to initiate fertility transition by increasing the supply of children may instead, as both Cleveland and Weil (this volume) demonstrate, reduce the perceived risk of shortened periods of postpartum infecundability and be used intentionally to increase the total fertility rate. Handwerker (this volume) argues that a perceived need for high fertility may be maintained in the face of sharp reductions in infant mortality by the mortality experiences of a small proportion of families. Increased per capita income may not stem from improved health, and increases in per capita income often do not, in fact, transform a situation of excess demand to a situation of excess supply. The findings of Ross, Alexander, Schumann, Odell, Whiteford, Cleveland, Weil, and Hayden (this volume) all suggest that per capita income may be a function of the family labor supply, and that new goods and services may be available only to couples with large families. As the papers by Whiteford, Stamm and Tsui, and Dorjahn (this volume) reveal, urbanization itself has no direct link with fertility; any macro-level correlation between urbanization and changes in fertility must reflect the operation of other variables. The modernization interpretation of fertility transition founders on the assumption made explicit in Easterlin's contributions that in all contexts children are net consumers. The studies in this volume argue, to the contrary, that in agrarian contexts children are net income producers. The relationship between education and fertility transition has been shown to be equally ambiguous (Cochrane 1979, Graff 1979). The findings presented in this volume confirm a regular association between education and fertility. The character of that association appears to vary, however. In Guatemala (Odell, this volume), high income families with low fertility can afford higher levels of education for their children, but the rationale for small families is not clearly or importantly tied to the increased costs entailed by schooling or to the educational levels of these couples. Rather, the rationale appears more closely tied to political considerations. In Ireland (O'Reilly, this volume), the school system itself indoctrinates for high fertility. In Tunisia (Stamm and Tsui, this volume), the impact of education is conditioned by perceived relationships between family size and adult status and prestige. In south Texas (Whiteford, this volume), high fertility levels appear to be the outcome of a lack of access to the economic opportunities made available by education.

Deficiencies in Standard Demographic Transition Theory Thus, despite both an extensive reconstruction of demographic transition theory, and the broad historical and contemporary inverse relationship

8

HANDWERKER

between fertility and economic and social "modernism," we still do not have a clear idea why or how the changes that have accompanied the emergence of a world industrial system have brought about fertility transition. I believe this state of affairs may be traced to two weaknesses in the modernization interpretation of fertility transition. The first is its failure to venture beyond the comfortable confmes of our own intellectual tradition to renounce the view that a process of "modernization" cum "Westernization" cum "development" exists, and its corollary that there are "stages" of these processes. Not only has an immense literature revealed these assumptions to be histQrically naive, the typology used in subsequent analyses, by its very nature, cannot comprehend the complexities documented by studies such as those in this volume. The second is its current grounding-or over-reliance-on economic decision-models of fertility. In point of fact, we face the problem that simple and rigorously stated microeconomic models which simultaneously are valid beyond particular regions and historical periods have been notoriously hard to uncover (Schultz 1977). In this volume, Crosbie points out that the linkage between rational decision-making and fertility generally has been imputed rather than demonstrated, and, once tested, turns out to be questionable even when, as in the contemporary United States, couples have ready access to effective modes of contraception and induced abortion. These two characteristics of standard transition theory mean that it must beg the question of how and why people should reconceptualize their relationship to . children in ways that lead them to sharply reduce their family size. The papers in this volume generally support Bongaarts' identification of the important intermediate fertility variables. Bongaarts' formal model is rarely used, however. First, the model cannot be used to estimate fertility. The accuracy of the model is determined largely by statistical dependence: TFR is calculated from indices dependent on the prior calculation of TFR. The model is properly used (see the papers by Cleveland and by Brainard and Overfield) only to estimate the relative importance of each of the four variables of which it is constituted. Second, as Wilmsen (this volume) suggests, dietary stress may interact with and unusually lengthen lactationinduced amenorrhea, or alter fecundity, and model estimates may be distorted. Third, the model limits itself to proximate fertility determinants, one of which, proportion of married women, is a proxy for more immediate determinants of the frequency of coitus relative to ovulation, including age at marriage, polygynous status, women's work loads, the mean duration of spousal separations, and the relationship between separations and fecundability, age differentials between marital partners, and divorce, mortality, and re-marriage rates. Finally, by limiting its focus to the proximate determinants of fertility the model begs the broader questions of the

CULTURE AND REPRODUCTION

9

determinants of intermediate variable parameters. It is this last question that we focus on in this book. The findings presented in the body of this volume do not, however, support Henry's distinction between natural and target fertility populations. Some findings might be taken to suggest that the concept of natural fertility is a useful heuristic. Despite the unfortunate choice of the valueladen term "natural," and varying total fertility levels, the differences in the age pattern of childbearing between populations which regularly and effectively practice contraception and induced abortion and those which do not is substantial and reasonably consistent. The prolonged childbearing among populations that do not sharply truncate reproductive careers is both socially and culturally significant. As both Schumann and Odell (this volume) point out, for instance, such childbearing raises health problems which do not otherwise exist. As the papers by Whiteford, O'Reilly, and Stamm and Tsui (this volume; cf. Scrimshaw 1978b) document, such childbearing circumscribes the activity options women have before them, sometimes severely. A frequently cited rationale for wanting, or for adopting, family planning technologies is an altered perspective of the extra-household activities that may be legitimately pursued by women, but which are blocked by having to stay home with children. Natural fertility does not, of course, mean the absence of fertility control. As Wilmsen (this volume) points out, the universal institution of marriage itself is a major control on fertility. Moreover, nearly all populations utilize means (however ineffective) for contracepting and inducing abortions (contrast Schumann's report in this volume). Although the use of these technologies normally is circumscribed (see Devereaux 1955), both Weiland Cleveland (this volume) show that "natural" fertility populations have and do intentionally and effectively either increase or decrease fertility. That Henry's distinction between natural and target fertility populations is not in fact a useful heuristic, however, is most dramatically revealed by the finding that the age pattern of childbearing characterizing natural fertility populations is consistent with the intentional truncation of reproductive careers using modern contraceptives. "Natural" fertility cannot be distinguished from "target" fertility when the target is a large number (see, especially but not exclusively, Brainard and Overfield's paper on Western Alaskan Eskimo). Ambiguity is inherent to the concept and is its most serious liability. As Butz (1982) points out, the concept of natural fertility remains a residual category that begs the question of its determinants. The findings of this volume reveal that natural fertility regimes consist of a discrete and identifiable set of determinants. Unless its determinants are specified for particular historical periods and conditions, informed policy and

10

HANDWERKER

intervention is impossible. Because, as Handwerker shows (this volume), natural fertility determinants can be modeled for specific populations, the concept itself becomes superfluous. The degree and kind of reproductive career truncation, as Polgar observes, reflects cultural specifications " ... concerning who should have children, when childbearing should start, what is a desirable interval between children, and what juncture in social aging childbearing should cease" (1972:209; emphasis in the original). From this perspective, conceptualizing fertility transition as a transformation of a natural fertility regime into a target fertility regime mis-specifies the issue. Fertility transition reflects a cultural transformation.

Culture and Reproduction This conclusion is consistent with Knodel and van de Walle's (1979) review of historical data from Western Europe, which found that the single "factor" to reliably influence the timing of fertility transition was "culture." Aside from mentioning that the status of women seemed important and pointing out that language tended to channel the diffusion of family planning information and knowledge, however, the contents of "culture" remained unspecified. They concluded that "The identification of social and cultural indicators that would reflect receptivity to family limitation... would be of considerable value to social scientists and policymakers alike. More effort needs to be made in this direction" (1979:238). As Ross (this volume) points out, however, too often "culture" -:-like "natural" fertility-is taken to be the black box of unknowns that merely expresses our ignorance of the determinants of reproductive patterns. Anthropologists' failure either to clearly identify the different useages of the concept or to thoroughly explore the implications of their own diverse approaches, appears to be responsible for this situation. Demographers would like to use the concept as an independent variable. Anthropologists, however, use the concept as a dependent variable (e.g., Ross, this volume) as well as an independent variable (e.g., Stamm and Tsui, this volume). When they assign the concept the role of independent variable, invariably and probably necessarily the concept reveals ambiguities and complexity-rather like the micro-demographic findings reported in the body of this volume-instead of tidy solutions that might fit into formal general models. When used as an independent variable, culture turns out not merely as all those factors other than economic, political, social, religious, and

CULTURE AND REPRODUCTION

11

psychological; culture is not distinct from these phenomena. Culture refers to the ideas by which individuals order material experience and assign value to its elements. Culture thus refers to the content and structural specifications that people use to define economic, political, social, religious, and psychological factors. As an independent variable, culture is defined as a mental, not a material phenomenon; as a phenomenon that is created and changed by people in their attempt to make comprehensible a material reality that does not compel us to conceptualize it in one and only one way-and thus requires that we theorize about its nature; and as a phenomenon whose change entails continuity with the experience of our individual and collective past. It follows from this perspective that variables included in decision-making models cannot have consistent effects. The effect of a given variable must be contingent on the manner in which it is conceptualized, and the manner in which it is conceptualized must be relative to specific culture regions and culture-historical periods. Because the essentials of culture (as an independent variable) consist of axiomatic assumptions about the nature of material reality, culture defines, and thereby limits, the values assumable by a specific variable, specifies the linkages among variables that will constitute the basis for behavior, and specifies the content and priorities of the cost-benefit calculi in terms of which people make decisions. Thus, "religious" factors admittedly can influence reproductive decisions and "Catholicism," as one form of religion, is cited often as an important determinant of fertility. Yet the studies in this volume by O'Reilly, Schumann, Odell, and Whiteford reveal that what constitutes "Catholicism" in Ireland is in some ways radically different from "Catholicism" in Latin America and that ·"Catholicism" is not the same variable for different portions of Latin American populations. Hence, attempts such as Easterlin's to find a simplifying decision-model of the determinants of the intermediate fertility variables are likely to prove futile. Such attempts must be based on the assumption that there is a negligible discrepancy between the model-maker's specification of the variables constituting the model and the understanding of those variables held by the population(s) to which the model applies. Such an approach unjustifiably and unrealistically reifies those variables. To judge from the findings presented in this book, these assumptions are false and lead inevitably to the common problem mentioned above: mis-specified m.odels that are poor policy tools at particular space-time loci (see Crosbie, this volume). Culture theory has a more important, and a largely unexamined, implication: decision-making models must be either mis-specijied or covert tautologies. Without cultural specification, the concept of a cost-benefit calculus will be vacuous, for choices can be irrational only for the person

12

HANDWERKER

observing the choices made by others, and then only when the observer starts from the wrong premise. It is this observation, of course, that underlies the importance of the concept of culture as an independent variable. Behavior that otherwise is aberrant is comprehensible once we find the premise from which the decision to perform that behavior follows logically. However, it also follows that when such models are either (i) constructed post hoc-as in the Geertzian "thick descriptions" anthropologists are wont to produce - or (ii) re-specified in light of local cultural constructions, such models will constitute covert tautologies. When one delineates the premises and strategi~s on which particular decisions to perform specific behavioral patterns are based, one must define those premises and strategies in terms of those decisions: behavioral option X was chosen over behavioral option Y because X was preferred over Y because X is more beneficial than Y in the specific cultural context C. Such statements must always be true no matter what the behavioral content of X and Y. This means (see below on Caldwell's model) that it is impossible to identify the model independent of the behavioral pattern it purports to explain. This isomorphism between ideas and the behavior they generate means that such "explanations" necessarily beg the question of the determinants of the ideas and behavioral patterns we wish to understand. If we wish to produce sophisticated descriptions of behavioral patterns, "culture" is appropriately used as an independent variable. However, if our goal is to account for the origins, maintenance, or change of behavioral patterns, the concept is appropriately used differently. Changes in behavioral patterns we wish to explain necessarily will correspond with changes in the mental templates that generate those patterns. The behavioral pattern and the ideas they presuppose jointly constitute a dependent variable. Because cultural changes necessarily exhibit continuity from one state to another, culture at time t also is used appropriately as a control variable specifying the point from which change must take place, and determining (at least in part) the trajectory of change from that point. Wealth Flows Theory: A Promising Alternative Wealth flows theory (Caldwell 1982) is a general theory of social and cultural transformation that, with suitable modifications, holds the potential to avoid the conceptual problems reviewed above, and to account for many of the empirical components of demographic transition that, from the perspective of the standard, modernization interpretation, remain puzzling. Caldwell argues that a great divide separates high and low fertility regimes; that this divide is a function of social structure and, more specifically, of

CULTURE AND REPRODUCTION

13

distinctive family moralities specifying the relative obligations of parents and children; that these family moralities dictate that in the former the net flow of wealth from children to parents (and more generally between generations) is positive, and that in the latter the net intergenerational wealth flow is negative; that, therefore, unlimited fertility is economically rational in the former and that zero fertility is economically rational in the latter; that where the intergenerational wealth flow is negative parents continue to have children, but only in small numbers and for non-economic reasons; that the transition from a high to a low fertility regime is defined by a shift in the intergenerational wealth flow but that this shift is determined by a change in family morality; and that this change in family morality occurs with the onset of mass schooling based on a Western model; hence, that fertility transition in current LDCs is a function of the diffusion of Western values through schooling and the mass media. If Caldwell is right, most of the standard indices of development or modernization (literacy rate, infant mortality, urbanization, proportion of the labor force employed outside the agricultural sector) are irrelevant and can be expected to display marked variation both prior to and during fertility transition. The single appropriate social indicator of the onset of fertility transition would be the existence of mass public education. Theoretical and Empirical Gaps in Wealth Flows Theory Caldwell's approach and claims have been questioned (e.g., Thadani 1978; Hawthorne 1978) on several grounds: (i) the empirical support he adduces is slim, (ii) he ignores the primacy of economic forces over social forces, (iii) he accounts neither for the absence of extremely high fertility in pre-transition Western Europe and the persistence of moderate rather than minimum fertility in the West for generations after the transition. More recently, Willis (1982) suggests that Caldwell has reversed the direction of causation between family change and economic development. In his 1982 summation, however, Caldwell observes that the first two ''criticisms'' are, in fact, empirical issues both he and others are evaluating, and the last issue reflects the discrepancy between the historical conditions under which the Western European and contemporary fertility transition take place. Willis (1982) addresses an issue peripheral to wealth flows theory and adduces no evidence in support of his alternative hypothesis. The fmdings of the present set of research reports adduce considerable support for wealth flows theory (e.g., the papers by Ross, Alexander, Cleveland, Weil, Schumann, Odell, Whiteford, Dorjahn, and Handwerker). Key claims of wealth flows theory have not, however, been submitted to

14

HANDWERKER

clear tests. Moreover, the theory itself contains two major gaps. First, Caldwell's hypothesis that " ... the attainment of universal [Westernized] schooling across a single nation or cultural group is the force that changes intergenerational attitudes and hence [intrafamily] economic relationships" (1980:321) has not been validated (Handwerker 1981b), ignores situations such as in Ireland where the Western school system indoctrinates for high fertility (O'Reilly, this volume), and begs the question of why values change. Second, Caldwell's claim that attitude change determines behavioral change is a covert tautology (Handwerker 1981a,b). When Caldwell argues that f~ily morality determines intergenerational wealth flows he implies the more fundamental proposition that what people do follows from how they order experience and assign value to its elements-the view in which culture is assigned the status as independent variable. A given description of family morality will, necessarily, correspond with a particular intergenerational wealth flow and to say otherwise would be self-contradictory. This correspondence necessarily follows because from the point of view of the people we wish to understand, any change in behavior (e.g., the direction of the intergenerational wealth flow) implies a change in the way they conceptualize experience and assign value to its elements. Conversely, a change in the way in which people conceptualize experience and assign value to its elements (a new family morality) implies a change in behavior (a new intergenerational wealth flow). In short, we cannot identify specific behavioral patterns and the ideas they presuppose independent of one another. To "explain" behavior by reference to those ideas therefore constitutes a covert tautology. FiiUng the Gaps: Modifications to Wealth Flows Theory These gaps can be filled by working from the premise that changes in material constraints alter values and behavioral patterns because they change the means by which people attempt to create and stabilize income flows at acceptable levels (Handwerker 1977, 1983). Such a change, I argue elsewhere (Handwerker 1983), marked the transition between foraging and agrarian societies. Such a change, I suggest, also marks the transition between agrarian and industrial societies (see Hayden's overview, this volume; Handwerker 1986). Caldwell (1982) has adduced a wealth of evidence supporting the view that among settled cultivators the wealth flows that accrue to the parental generation either directly from children, or indirectly through the social relationships created by or through the use of children, contribute in important ways to patental material well-being. Even where such wealth flows

CULTURE AND REPRODUCTION

15

diminish relative to other sourc~s of income, there is still no reason to reduce fertility. Incentives to reduce fertility presuppose, as Caldwell (1982) observes, that this intergenerational wealth flow is reversed. When the perceived costs of children exceed the expected returns (in money, goods, and services), bearing children adversely affects parental aspirations for themselves (individually or collectively), parental aspirations for their children, or the material well-being of both in some combination. Simultaneously, this change in the economic structure of family relationships initiates the realignment in moral relationships that Caldwell identifies: an emotional nucleation in which the conjugal bond takes precedence over all relationships external to the nuclear family, an economic isolation of the nuclear from the extended family, and an emphasis on what parents owe children rather than on what children owe parents. As these moral transformations occur, fertility transition is initiated. As Birdwell-Pheasant observes of a village in northern Belize: Control over labor is no longer the aim of economic strategies; control of money has become paramount. Preliminary indications also suggest that Ramonaleiios have made a critical shift in evaluations in another area as well: children are no longer wealth, but expense. Ramonaleiios can see the way young families with money lavish their two or three children with clothes, jewelry, toys, etc. Large families, even with the increased wealth, are unable to match such per-child expenditures. So the women begin to talk of having fewer children. A few women have, by all indications, even begun to utilize contraceptives (1984:12).

The industrial system brings with it a revolution in the means by which material well-being is best created and maintained: kinship and other personal relationships are superceded in importance by alternative lines of access to strategic resources. To the extent that material well-being can be created and maintained through personal relationships, children will be important sources of income flows. To the extent that material well-being must be created on the basis of some other criterion, the income flows accessible through such social relationships must diminish. High fertility comes to adversely affect the material aspirations of parents, for themselves, their children, and/or both; the moral economy of reproduction and family relationships changes, and fertility transition is initiated. The relative importance of kinship and other personal relationships can change only when alternative lines of access to strategic resources proliferate. Formal education and skill training, which become of widespread importance only in the context of the industrial system, offers for the first time in human history a criterion that can be used independent of kin and

16

HANDWERKER

other social relationships to gain access to strategic resources. However, such proliferation cannot occur unless this novel line of access to key resources yields greater returns more reliably than the use of personal relationships. Educationally-acquired skills and experiences can yield such returns only with changes in opportunity structure and the labor market. Such changes in opportunity structure and the labor market have the effect of constraining the income flows accessible solely through children and the personal relationships (to kin, friends, and patrons) they can create. Hence, fertility transition should not follow from the onset of mass education, as Caldwell claims, but from the conjunction of mass education with changes in opportunity structure that increasingly reward educationally-acquired skills and perspectives. These events can, of course, be manipulated by government, as recently in Singapore and China. However, historically these events have come together primarily with economic growth and the structural differentiation of local economies. The conjunction of these events can be expected to bring about marked improvements in the material welfare of populations, and to both initiate and influence the pace of fertility transition. Regional, social class, and historical variation in the conjunction of these events would lead to regional, social class, and historical variation in the initiation and pace of fertility transition. This argument is consistent with findings (e.g., Faruqee 1979) that improved standards of living correlate with fertility transition, but this argument is not that improved standards of living bring about that transition. My argument is that fertility transition is brought about by (1) changes in the moral economy of reproduction and parent-child relationships, which change because (2) the expected wealth flows from children and the social relationships they create is negative, which occurs when (3) the relative importance of personal relationships in accessing key resources declines, which declines because (4) changes in opportunity structure are accompanied by the ability to make effective use of an independent criterion-most commonly but not exclusively, formal schooling and skill training-to create and maintain acceptable standards of income, which occurs primarily when (5) competitive non-agricultural economic oppor· tunities proliferate. Once begun, fertility transition appears irreversable (cf. Knodel and van de Walle 1979), for we have never documented a counter change to a situation in which social relationships, independent of formal education and skill training, become more important than the latter in determining one's personal material welfare. Relative prosperity has independent effects that cannot initiate transition because those effects bypass the pertinent moralities of childbearing and parent-child relation· ships. However, those effects can influence the pace of a population's

CULTURE AND REPRODUCTION

17

fertility decline, hastening that decline if relative prosperity is low, or retarding that decline if relative prosperity is high. Education and Fertility This line of reasoning leads to a reconceptualization of the role of conventional socioeconomic indicators of the onset and pace of fertility transition. The inverse relationship between fertility and education, for instance, is the most consistently confirmed correlation in the literature on fertility transition (Graff 1979, Cochrane 1979). Speculations about why that correlation exists are among the most diverse in the literature. Education has been presumed to bring about fertility transition because (i) educated people are more willing to change their behavior, (ii) educational processes "break down" or "weaken" traditions and customs, (iii) education creates a "modern" and more rational perspective on the world, (iv) education makes one aware of greater range of opportunities and options for behavior, (v) education today is normally based on a "western" model and thereby serves to "westernize" people who pass through the system, (vi) educated people have greater access to (and lower costs for) effective forms of contraception and abortion, (vii) for women, education increases the opportunity costs of having children, (viii) education creates new tastes and preferences, (ix) education creates increased child-rearing costs. Of course, these speculations are not mutually exclusive, and are combined in various ways in the work of different authors. These claims beg the question of why and how "education" can have its reputed effect on fertility. In some cases (e.g., the "modernization" and "westernization" interpretations), this fallacy is concealed beneath a covert or overt tautology: "modernism" or "westernization" is defined by reference to declining fertility (see, e.g., Caldwell1982:287-9, Handwerker 1981a,b). In still other cases, this fallacy is concealed beneath an otherwise plausible claim: for instance, that education makes one aware of a greater range of opportunities and options for behavior, including the option to sharply limit fertility, the awareness of the costs of high fertility (especially for women), and the means for safely achieving this end (iv, and vi-ix, above). Since this formulation presupposes choice, however, the underlying question is why people would follow this line of reasoning and make the set of choices leading to reduced fertility? To take for granted that they would do so, as Caldwell (1982: 83-112) has observed, is ethnocentric and both unrealistically and unjustifiably presumes a universal preference function. There are ample grounds for working with the assumption that people tend to maximize the amount and reliability of their personal income flows, but

18

1

HANDWERKER

both the definition of income flows and the means by which this goal can be best achieved vary widely. Given this state of affairs, it is not surprising to find "Skepticism [about the reputed influence of education on fertility] turn to doubt when we find that changing levels of education and/ or literacy seem to lead sometimes to lower fertility and at other times to increasing birth rates" (Graff 1979: 107). Graff suggests that " ... education, usually conceptualized and specified as an independent and quite significant variable, may in fact play a more indirect, mediating role in actual behavioral circumstances" (1979: 107). I agree, and argue that ed~cation or literacy by itself can have no important effect on fertility. The linkage between education and fertility is contingent on opportunity structure, and will turn on the issue of how material wellbeing can best be created and maintained, and how educationally acquired skills and perspectives fit, or do not fit, into this process. 1 Where a population cannot create and maintain acceptable levels of material welfare by using educationally-acquired skills and perspectives, or where such employment opportunities are limited, high fertility remains an important strategy for both creating and maintaining personal material well-being. Among Mexican-Americans in south Texas, for instance, ·those who have had both limited access to· education and limited access to employment opportunities based on skills acquired through formal schooling or skill training continue to have large families; those who have been able to make use of these skills in recent years are the people who have sharply limited their family size (Whiteford, this volume). These conditions are commonly met in cities throughout tropical Africa (e.g., for Liberia, see Handwerker 1987). Employment opportunities are dominated. by government. Government employment depends heavily on personal relationships. Government employees are morally obligated, as well as self-motivated, to provide a variety of services and monetary assistance to parents, siblings, and other clients, both kin and friends. This assistance takes the form of personal appointments to open positions, the creation of jobs where none formerly existed, inside information on open positions, job recommendations, assistance with obtaining development loans and goods (new seed varieties, improved tree stocks), and lucrative government contracts. For large segments of urban populations, the centralization of resources represented by the dominance of public sector employment, and the greater growth in public rather than private sector, means that children and the social relationships created by and through children continue to be perceived as primary determinants of individual material welfare. Aggregate fertility thus remains high even in urban centers where surface changes in opportunity structures might otherwise lead one to expect at least the initial stages of fertility transition.

CULTURE AND REPRODUCTION

19

Conversely, where changes in opportunity structure and the labor market increasingly reward educational skills and training and thereby increasingly constrain income flows through social relationships, it becomes increasingly probable that parental material welfare and aspirations are adversely affected by high fertility. This contingency is clearly delineated in a detailed ethnohistorical study of social class-specific fertility transitions in a Sicilian town (Schneider and Schneider 1984a,b). Fertility transition among the civile class-the landed gentry-was initiated earliest, in the 1880s, when unlike peasants and artisans, the former was no longer able to generate important income flows from or through their children. The availability of cheap, machine-produced North American grain led to a collapse of grain prices, and large-scale emigration depleted the agricultural labor force. These changes in opportunity structure meant that the welfare of civile children increasingly required heavy investments in education that could not be recouped" ... at a time when the land no longer yielded generous revenues and the labor force was depleted" (1984a:256). High fertility, consequently, became increasingly burdensome. During this same period, on the other hand, peasants and artisan families participated heavily in out-migration to the United States, and reaped the benefits of emigrant remittances, improved local labor contracts, and an export market for Sicilian foods. Fertility transition among artisans was initiated in the 1920s. By that period, emigration had become sharply restricted, the perceived wealth flows from or through children had disappeared, and educational costs (for apprenticeships) " ... became an increasing burden, the more so as the revaluation of the lira and the world depression kept clients from paying their bills" (1984a:260). High fertility came to be viewed as detrimental to the individual material welfare of both parents and children. Fertility transition among landless peasants was not initiated until the 1950s and 1960s. Prior to this period, this bracciante class had sharply limited employment opportunities. As the Schneiders point out, "In the prewar period, for a man to be hired as a field hand he might also have had to commit the labor of his wife and daughters to the landowner's household, just as he and his sons were expected to be 'on call' for various work prestations and errands, ... " (1984b:6). In the absence of employment alternatives, and where both the probability and security of employment depended on children, there was no incentive to reduce fertility. Such incentives did not come into being until agrarian reform in the early 1950s and the emergence of contractual wage labor created opportunities for the braccianti to free themselves of servitude on the great estates. Only in the presence of this change in opportunity structure was high fertility perceived as burdensome and "a better life" made possible by reducing the size of

20

HANDWERKER

families. Hence, fertility transition does not follow merely from the onset of mass education. On the contrary, because parents who are limiting their fertility are those who conceptualize their responsibilities to their children to include preparation for the new opportunity structure, the pace of fertility transition is likely to influence the pace with which public education, especially secondary education, becomes prevalent in a population. Infant Mortality and Fertility The modern demographic transistion, of course, consists of both a fertility transition and a mortality transition. The view that high fertility is a response to high mortality, that fertility transition should follow from mortality transition, and that both follow from "development," is central to the modernization interpretation of demographic transition theory (Notestein 1945, Davis 1963, Heer 1983). Historical population trends in Europe (McKeown 1976) gave rise to these expectations, and population trends among current Less Developed Countries between 1950 and 1980 (Mauldin 1980) give them further face validation. Although precise specifications often are elusive (Chen 1983), mortality decline is widely ascribed to the conjunction of three factors: (i) improving maternal and child nutrition, the outgrowth of an increasingly efficient world food production and distribution system; (ii) improving access to medical care, and (iii) new public knowledge about the determinants of maternal and child health. The increasing prevalence of public education is regularly linked with factors (ii) and (iii) (Caldwell1979, DaVanzo, Butz and Habicht 1983, Martin et al. 1983). The presumed links between declines in infant mortality and fertility transition, however, have been elusive. Close inspection has revealed that the linkages among these phenomena are not consistent from one region and time period to another, and frequently are not known. After explicitly probing the qualitative dimensions of Thailand's fertility transition, for instance, Knodel, Havanon, and Pramualratana (1983) found that "While there was close to universal recognition among participants that survival chances of children had improved considerably over the last few generations, only occasionally were these mortality risk differences mentioned in connection with the intergenerational differences in family size .... Given its low saliency, ... , we see little basis for believing that the reduction in mortality [in Thailand] acted as a factor to precipitate fertility decline" (1983:66-67). Taylor et al.'s (1976) conclusion that improved child survival is neither necessary nor sufficient for fertility transition is amply justified by the

CULTURE AND REPRODUCTION

21

available data. High fertility, in fact, may not be a response to high mortality. Scrimshaw (1978) has shown, to the contrary, that under some circumstances high mortality (the outcome of overt or disguised infanticide) is a response to high fertility. In a large number of cases, "development" and declining mortality has been accompanied by fertility increases (e.g., Nag 1980). Two such cases that have been examined in detail involve Kusasi intensive cultivators in northern Ghana (Cleveland, this volume) and Mandinka farmers in The Gambi (Weil, this volume), and their response to the declining risk of infant mortality is instructive. Cleveland examines the changing determinants of fertility for the Kusasi over a 50 year period. He demonstrates that extensive emigration and land degradation increased Kusasi demand for labor. Changes in marriage institutions have led to significant reductions in the mean age of first marriage over this period. However, Cleveland argues that Kusasi fertility increases have come primarily from an intentional reduction in the period of postpartum abstinence. Kusasi couples have been making decisions to decrease birth intervals because improved child health has allowed them to do so without decreasing survivorship. Weil traces the intensive agriculture now characteristic of much of The Gambia from its origins in 19th century political economy, and shows how specific changes in technology have altered the division of labor and the composition of Mandinka householdproduction units. As such units have become increasingly dependent on their own reproduction to supply labor, they have increased the supply of labor by intentionally ignoring traditional prescriptions concerning lactation and post-partum abstinence. Children, either directly or through the social relationships that they create or can be used to create, remain central to the perceived material well-being of these people and fertility remains high. Although declining mortality may be neither necessary nor sufficient for fertility transition, perhaps the risk of mortality influences the pace of transition? The line of reasoning developed above, however, suggests that high infant mortality provides support for high fertility - as in contemporary Liberia (Handwerker, this volume) - only where parents depend upon children to access important resources. Failure to recognize this contingency leads investigators to the unjustified conclusion that the linkage between mortality and fertility persists after the initiation of fertility transition. As Knodel, Havanon, and Pramualratana observe in their evaluation of Thai data, '' .. .it seems intuitively likely that had mortality risks remained at the levels prevailing several generations ago, the present fertility decline would have been more modest" (1983:66-67). Where children are no longer-either directly, or indirectly through the social relationships they

22

HANDWERKER

create or can be used to create-important productive assets, infant mortality becomes largely dissociated with fertility. Indeed, even in agrarian communities in times of food scarcity, a new birth is likely either to die or to be killed. As Scrimshaw (1978) documents, infanticide tends to be widely practiced, but irregularly and rarely intensively, as a response to recurrent instances of resource stress. Because fertility transition presupposes a re-evaluation of the role of children that dissociates fertility from mortality, fertility may begin to decline before the onset of mortality transition, the onset of mortality transition will not influence ~he timing of the onset of fertility transition, and the pace of the infant mortality transition will not influence the pace of fertility tansition. 2 On the contrary, the onset of fertility transition and the pace with which it proceeds correlates with the pace of mortality transition because parents who are limiting their fertility will be those whose educational and family income levels facilitate their access to improved infant nutrition and health care.

Family Planning Programs and FertiUty In 1980, W. Parker Mauldin expressed what he felt was an emerging consensus in concluding a review of world population trends and prospects with the observation that Family planning programs have a significant, independent effect [on fertility], certainly in developing countries with favorable social settings and also under certain conditions in countries with less favorable settings (including the three largest: China, India, and Indonesia). Moreover, the longer the program and the clearer its demographic intent, the greater its effect (1980:155).

Such a conclusion, I suggest, lacks both theoretical and empirical grounds. The difficulties in "motivating" people to make use of family planning services are legendary. Even when some family planning services are adopted, fertility transition has not always followed. As Caldwell and Ware observe for Nigeria: "Even the educated are adopting contraception largely in order to retain existing birth intervals by substituting sexual activity controlled by modern contraception for the traditional period of postnatal abstinence, ... " (1977:507). Yet, fertility transition has occurred historically in the absence of modern contraceptives, and some populations (e.g., the bracciante cited above; Schneider and Schneider 1984b) have rejected modern contraceptives as the means by which to accomplish fertility transition. Few people would quarrel with the view that family planning

CULTURE AND REPRODUCTION

23

services that are poorly advertised, costly, and difficult to use, will receive little use, or that the availability of well-advertised, inexpensive, and easily used family planning services can influence the pace of fertility transition (Retherford and Palmore 1983). If the line of reasoning developed in this paper is correct, however, family planning services-like education and mortality transition-cannot initiate fertility transition. The adoption of family planning services for the purpose of sharply limiting fertility assumes the existence of new, attractive options-the emergence of new opportunity structures (Davis 1977, Caldwell 1982: 118-33, 272-90) and the means to access those structures independent of social relationships. In short, the pace at which people accept family planning services for purposes of sharply limiting fertility is dependent on the pace they can access new opportunity structures independent of social relationships. Where the emergence of new opportunities has been slowed, the pace of acceptance will slow. Where such opportunities remain limited, no amount of advertising, cost reductions, exposure to the mass media, "westernization" of education, or proliferation of family planning resources, can initiate fertility transition. Even if family planning technologies are adopted, they will be used not to initiate fertility transition but, as among Western Alaskan Eskimo (Brainard and Overfield, this volume), to achieve control over a supply of children that remains at a very high level, or as in Liberia (Handwerker, this volume), to avert pregnancies when women are in school, and abandoned once schooling is completed. If new opportunity structures that increasingly reward skills and perspectives developed in formal educational systems come into being, however, fertility transition will be initiated. This contingency is revealed in Nag and Kak's (1984) restudy of Manupur, the Indian village made famous by Mamdani's The Myth of Population Control (1972). Nag and Kak's research documented a widespread reversal of attitudes toward family planning which had been brought about by changes in opportunity structure and the spread of public education which had only begun at the time Mamdani undertook his study. If family planning services do not exist, people will devise means-even means that require sacrifice and much selfdiscipline like abstinence and coitus interruptus (Schneider and Schneider 1984a,b)-to reduce fertility. They may, as in the Irish case (O'Reilly, this volume), illegally and under moral sanction (cf. Schumann, this volume), import contraceptives or move elsewhere to have abortions.

Reintegrating Demographic Theory with Evolutionary Theory Fertility transition, like many other social and cultural phenomena, is a

24

HANDWERKER

micro-level process with both macro- and micro-level determinants. Its micro-level determinants include changes in the intermediate fertility variables and cultural change in the manner in which these variables are conceptualized and structured relative to one another. Criticisms of the modernization interpretation of demographic transition, especially those based on the findings of the Princeton European Fertility Project, have led in recent years to a search for culturally-based mechanisms (e.g., Knodel and van de Walle 1979) of fertility transition that capture the divergent culture-historical conditions bearing on fertility. Caldwell's emphasis on Westernization is one suclt effort. Unfortunately, treating culture as a conventional independent variable begs the question of the interactive relationship between culture and empirical conditions through which cultural changes come about. To address this question, we need to re-address macro-level issues, for changes in macro-level constraints seem to be the determinants of the micro-level processes through which fertility transition works itself out. This is the case, I suggest, because culture changes reflect a selection of novel conceptual structures and the behaviors they generate based on their empirical effects. We have been trying to answer questions raised by the modern demographic transition with strategies that by their very nature cannot provide the answers. The problem of demographic transition is one of accounting for the change of system states. The primary tools have come to be micro-level investigations and modeling of decision making. The explanations we have generated all take the form: (1) if people make assumptions AI, Az, ... , "' (2) within the set of constraints ci' c2, ... , en, (3) they will conform to the behavioral pattern B. Such models are useful for describing equilibrium states (e.g., the rationale for high or low fertility). But the implication of such a focus is that if we only increase the sophistication of our decision-making models we shall be able to account for demographic phenomena. In fact, I argue, micro-level investigations can only produce more sophisticated descriptions of phenomena that still need to be accounted for. Whether formal optimization models or informal descriptions of cultural context and decision making, such models can only generate behavior for the constraints defining the system state; "explanation" of B by reference to stipulated assumptions and constraints is necessarily a covert tautology. Transition from one state to another only comes about through changes in assumptions or constraint specifications. Since the behavioral pattern will be a necessary logical consequence of model assumptions and constraints, quantum improvements in understanding can come about only when we are able to account for those assumptions and constraints. Change in constraint specifications is a macro-level phenomenon. Two

CULTURE AND REPRODUCTION

25

realms of theory that apply to macro-level phenomena are "modernization" theory and cultural materialism. Modernization theory, however, does not conceptualize constraint specifications because it is formulated as a typological theory and hence substitutes a description of differences between states (Traditional and Modern, Natural and Target Fertility) for a mechanism of change. Perhaps standard theory persists in its present form because we are too wedded to the idea that the concept of social or demographic "type" is of central importance. We balk at making a self-contradictory claim: that the concept of type is superfluous. The evidence presented in this volume suggests that, if the central puzzle of demographic transition is to be solved, we must make this claim. If we persist in assuming that such types are something other than mere epiphenomena, paradigm shifts in demography will continue to constitute quibbles about the specification of system states. We will be unable to solve the central puzzle: perhaps to stimulate fertility transition we only need more modernization? But what, among all that the term "modern" encompasses, do we need more of, and why? Analyses produced from the framework of cultural materialism do not suffer the deficiencies of standard demographic theory because they assume that "types" are epiphenomena and they look to change in constraint specifications for the mechanisms of cultural change. However, cultural materialism has not moved beyond the application of ambiguous general principles to generate an encompassing theory that could integrate what now stand as idiosyncratic studies. Evolutionary theory is designed to accomplish this specific end and thus offers the potential to generate quantum improvements in our understanding of human reproductive behavior. There remains the problem of producing a general theory of evolution unifying the biological and cultural dimensions of Homo sapiens.3 Such a theory requires answers to two central questions: (1) on what assumption or assumptions do people generate behavior, and why? and (2) how and why do changes in constraint specifications come about? The most common model assumption is that people are optimizing creatures. But what do we optimize, and why? Do people consistently optimize one and only one variable, a small number of variables, or a wide range of variables? Perhaps cultural diversity is explained by the optimization of a wide range of variables? Alternatively, perhaps cultural diversity is explained by the optimization of one or a small range of variables in the context of constraints that are unique to the life history of the individual or the population? In either event, however, it would seem that patterned human behavior would be explained by the optimization of one or a small range of variables subject to consistent constraints. Studies in this volume and elsewhere consistently reveal that material

26

HANDWERKER

variables bearing on our expenditures and acquisition of energy and income tend to have clear empirical referents posing important dilemmas for individual well-being. If people optimize some aspect of their relationship to resources, preference structures, decision-making processes, and the behavioral patterns to which they give rise, should-and to judge by the studies in this volume, do-correspond with differences in material context that bear on the acquisition and expenditure of energy and income. Moreover, changes in the parameters of such variables should -and to judge by the studies in this volume, do-regularly predict changes in culture and behavior. This finding provides grounds for concluding that changes in material constraints alter values and behavioral patterns because they change the empirical effects of an individual's attempt to optimize the amount, the reliability, and/ or their control of their personal income flows. All life forms require regular inputs of energy and nutrients. Ceteris paribus, selection would favor biological and behavioral propensities to control access to resources. The line of reasoning developed in the body of this chapter is a logical extension of this premise. To the extent that the creation and maintenance of material well-being is contingent on children and the personal relationships (with kin, friends, and patrons) children can be used to create, high fertility would be expected because it has positive empirical effects. To the extent that improvements in material well-being are contingent on the use of some other criterion, the income flows accessible through such social relationships must diminish. High fertility comes to have adverse effects on the material aspirations of parents, for themselves, their children, and/ or both; the moral economy of reproduction and parent-child relationships changes, and fertility transition is initiated. The essays in this volume do not explore the second question central to the development of a theory encompassing both biological and cultural evolution: how and why do constraint specifications change? The importance of that question is not, thereby, diminished. Is it that change in constraint specifications is a random (historically unique) process, or is change in constraint specifications non-random (perhaps even directional), determined by prior population, economic, or political states? Innovation and individual decision-making contribute to the creation of phenomena that have never before existed. What is the role of innovation and individual decision-making as determinants of macro-level change? Can we develop models that trace out macro-level determinants of micro-level processes, whose effect is to alter macro-level constraints? Evaluating these and other possibilities requires that we clarify interdependencies between biological and cultural phenomena and between biological and cultural evolution. What, specifically, are the linkages between our genome and cultural behavior and how do they operate? How,

CULTURE AND REPRODUCTION

27

specifically, does selection operate, at what organizational levels, and with what effects? Does fertility transition, for instance, represent the misapplication of evolved behavioral characteristics, as Irons (1977) suggests, or do we need to recognize a hierarchy of selection effects in which sharply truncated reproductive careers contribute directly to fitness only at higher levels of biological organization? Answering these questions implies a major shift in research strategy. We will need to devote more attention to the manner in which macro- and micro-level phenomena articulate, clarify the ways in which micro-level processes generate changes in macro-level constraints, and specify more precisely the manner in which those constraints shape behavior at the level of the individual. The outcome will not only be a fuller understanding of demographic transition, but also a fuller understanding of our place in the natural history of life on earth. Notes Acknowledgements. I thank Ted Ruprecht, Paul Crosbie, Todd Young, Peter Schneider, Donna Birdwell-Pheasant, Peter Weil, Ruth Kornfield, and Norman Schwartz for helpful suggestions and criticisms of earlier formulations of this argument. I have not had the opportunity to respond to all of their comments here, but hope to in the future. I thank Emmanuel N'Saingbe for initially guiding me into the path that led to this argument. 10ne should note that the specific levels of education that are linked with fertility, and the nature of their linkages, exhibit both historical and cross-sectional variation. Employment requirements, for instance, have changed over time, and vary regionally. In the case of the villagers of Ramonal in northern Belize mentioned above, formal education itself was relatively unimportant except insofar as educational experiences facilitated Ramonalenos effective movement into the commercial production of marijuana. Changes accompanying the First Industrial Revolution often displaced skilled workers and increased the relative contributions of children (Oshima 1983:602; Ross, this volume). More generally today, however, as Oshima suggests with data on the fertility transition in East Asia (1983), the increasing mechanization of technology has meant that secondary levels of education, rather than primary, are more closely tied with fertility transition. This historical change in the relationship between increasing employment requirements and technological complexity is apparent in time-series for Western European and North American countries. Thus, for instance, "The United States entered industrial society in 1900 with the steam-driven technology of the 'First' Industrial Revolution ... The decline in the birth rate was gradual until the 1920s, when there was a rapid spread of electric-and gas-driven technologies replacing unskilled labor, and an acceleration in secondary school enrollment" (Oshima 1983:600).

28

HANDWERKER

2one might object that fertility transition may be initiated when the conjunction of increasing fertility and decreasing infant mortality creates a burdensome number of children. The question that must be answered, however, is why those children would be "burdensome." The answer to this question is provided by the determinants of the importance of an infant's death: variations in opportunity structure that reward social relationships or skills and perspectives acquired in a formal educational setting. 3Criticisms of the logic and methodology of some claims made in the name of an evolutionary theory of culture (e.g., Handwerker and Crosbie 1982, 1983) do not imply, as for instance Turke (1984) believes, that such a theory is undesirable. Such criticisms imply only that tbe claims constituting such a theory be logically and methodologically sound.

PART ONE

Micro-Level Issues

2

Rationality and Models of Reproductive Decision-Making Paul V. Crosbie

Introduction

This paper reviews the case for rationality in reproductive decisionmaking. Students of human reproduction have increasingly come to reeognize the importance of decision and choice in the reproductive process. This recognition has fostered the formulation of a number of decisionmaking models of reproduction. Some of these decision models include assumptions of rationality in their frameworks. These rationality models view reproductive behavior as purposeful, as intended to yield some preferred state or set of consequences. As yet, however, tests have yielded only poor empirical support for these models. On these and other grounds, rationality models have received a great deal of criticism in the literature. The critics argue that reproductive behavior is not so purposeful or calculated as the proponents of rationality suggest. To date, however, these critics' arguments have not been accompanied by supportive data, and their criticisms seem to have gone unheard. This paper brings some data to bear on these issues. I find little evidence that rationality models predict more than the demand for children. I find clear evidence that alternative variables predict reproductive behavior. I infer that if we are to better understand the process of human reproduction we shall have to make major changes in current models of reproductive decision-making. 30

RATIONALITY AND MODELS

31

Rationality Arguments

The argument that humans are rational beings seems to have originated in the philosophical writings of Adam Smith, Thomas Hobbes and Jeremy Bentham (Lee 1971). These early philosophers suggested that humans act by anticipating the consequences of their decisions, and that humans would act best if they directed their decisions toward achieving the maximum consequential satisfaction or happiness. The past century has witnessed numerous elaborations of this argument and has seen the argument transformed from hypothesis to axiom to a national norm in the contemporary United States. Today, we find a variety of rationality explanations in the literature. Although these explanations differ in some respects, they tend to share a common set of assumptions. Briefly, these assumptions are that people seek or have information about the possible consequences of their decisions, that they weigh their decisions on the bases of these anticipated consequences, that they come to prefer those decisions which they think will provide them with the maximum consequential satisfaction, and that these preferences become the sole determinants of their decisions and subsequent behavior. In essence, then, the contemporary rational man (or woman) is informed and contemplates his or her decisions, and selects those decisions thought to bring him or her the greatest satisfaction. The assumptions of rationality were first formally incorporated into the explanation of reproductive decision-making by microeconomists. These microeconomists conceptualized reproductive decision-making as a form of household behavior which, along with the production and consumption of

other goods and services, would be governed by the economics of consumer choice. Microeconomic Theory

Microeconomic theory holds that consumers are rational in their decision-making and, as such, will allocate their limited resources in ways that maximize total satisfaction. The point of maximization or equilibrium is reached for a consumer when his or her demand or preference for an additional (price divisible) unit of any one commodity is equal to the demand for an additional unit of any other commodity. The principal determinants of demand or preference are income, tastes and prices. For most commodities, demand is assumed to be negatively related to the price of the commodity itself and is assumed to be positively related to income, taste (intensity of desire) for the commodity, and the prices of other commodities.

32

CROSBIE

Under this theory, then, the rational consumer will increase his/her demand and consumption of some commodity as the price of that commodity decreases or as income, taste for the commodity or the prices of other commodities increase (Boulding 1966; Gould and Ferguson 1980). Becker (1960) was the first to apply this perspective to the problem of reproductive decision-making. Becker argued that children were a household commodity and that the desire for children should be subject to the economic laws of demand. More specifically, he argued that households would be rational in their reproductive decision-making, and thus, the demand for children and reproduction would increase as the price of children decreased or as household income, the taste for children, or the prices of other household goods increased. Becker and others (Becker 1965; 1976; 1981:93-112; Becker and Lewis 1973; DeTray 1973; Willis 1973) later modified this argument to fit the "new home economics" conceptualization of consumer behavior. Under this conceptualization, the value of human time to be used in the production and consumption of household goods becomes, together with income, taste and prices, an additional determinant of the demand for those goods. As the value of human time (in terms of income producing potential) increases, the motivation to allocate time to the production and consumption of household goods will decrease, and the demand for those goods will decrease. As applied to the reproductive decision, Becker and his colleagues assume that only the wife's time is relevant to the production of children; husbands are assumed not to contribute to childrearing. Thus, in addition to the initial assumptions of income, taste and price effects on the demand for children, B~cker's revised model also assumes that as the value of wife's time increases, the demand for children will decrease. Becker and his colleagues have devoted more of their time to formalizing this model than they have to its empirical confirmation. As a consequence, no attempt has been made to test the full model or even to operationalize all of its terms. A number of attempts have been made to examine the relation between income and reproduction (family size), and these have produced inconsistent and inconclusive results (cf. Blake 1968; DeTray 1973; Gardner 1973; Michael 1973; Simon 1974; Willis 1973). Several attempts have also been made to examine the relation between value of wife's time and reproduction (cf. Cain and Dooley 1976; DeTray 1973; Gardner 1973; Michael 1973; Mincer 1963; Willis 1973). These studies, using wife's labor force participation and level of education as proxies for value of time, have found consistent, although slight, negative effects on family size. Easterlin (1966, 1968, 1969, 1973, 1975, 1978, et al. 1980) has incorporated the microeconomic framework into a broader, socioeconomic

RATIONALITY AND MODELS

33

theory of reproductive decision-making and fertility. While accepting the basic assumptions of microeconomic theory (i.e., the assumptions relating income, taste, and prices to the demand for children), Easterlin has argued that these assumptions are insufficient for a general theory of reproduction. He argues that a more complete understanding must include the determinants of taste for children and must incorporate fertility regulation and birth and death functions as explanatory variables. The birth and death functions in Easterlin's theory are treated as largely independent of rationality and reproductive decision-making and will not be further considered here. The significant additions to Easterlin's theory, then, are the arguments surrounding the determinants of taste for children and fertility regulations. Easterlin argues that a couple's taste for children is largely determined by the consumption behavior of their parents and parental generation. Since parents' education, income, social status, and religion are likely to have shaped their consumption behavior, then these (lagged) factors should be considered as determinants of their adult children's taste for children. Easterlin has made a particular case for the effects of parental income on the taste and demand for children. According to this argument, the higher the parents' income, the higher will be their adult children's aspirations for material goods. These higher aspirations for material goods will reduce the adult children's relative taste for children and, therefore, reduce their demand for children. Fertility regulation becomes an active variable in Easterlin's theory when a couple has reached their desired level of fertility and perceives that they could have additional children. At this point, they must adopt fertility regulating practices (abstinence, contraception, abortion or infanticide) or risk excess/unwanted births. Easterlin has suggested two determinants of fertility regulation; these are market cost (the effort spent to learn and use a practice) and psychic cost or taste for fertility regulation. "I:"he more nearly these costs approach zero (the perfect contraceptive society), the more likely a couple is to adopt fertility regulating practices and to limit their family size to the number of children desired. Like Becker, Easterlin has been more concerned with the formal development of his theory than with its testing. Indeed, the full theory is so abstracted at this point that it appears to be incapable of falsification (Sanderson, 1980). Nevertheless, one argument of the theory has received considerable empirical attention. This is the "relative economic status" hypothesis. This hypothesis is concerned with the combined influences of own (couple) income and parental income on the demand for children. According to this argument, as couple income increases (providing an expansion of the budget line), the demand for all goods, including children,

34

CROSBIE

will increase. On the other hand, as parental income increases, the aspirations or tastes for material goods will rise, these aspirations will reduce the couple's relative taste for children and, therefore, reduce their demand for children. Thus, the effects of couple income on the demand for children must be considered in context with and relative to the effects of parental income. Easterlin asserts that this relative income or relative economic status will be positively related to family size. A number of attempts have been made to test the relative economic status hypothesis, and these have produced conflicting and confusing results (cf. Chaudhury 1977; Crimmins-Gardner and Ewer 1978; Easterlin 1968, 1973, 1978; Ermisch 1979; MacDonald and Rindfuss 1978, 1981; Pullum 1980; Smith 1981; Sweezy 1971; Thornton 1978, 1979; Wachter 1975). A large part of the conflict and confusion in the results is doubtless due to the different measures which have been used as proxies for income. Many of these measures, such as cohort size, unemployment rate, and subjective wellbeing, are only tangentially related to income and are likely to include significant measurement error. Oppenheimer (1976) has identified another measurement problem in much of this research-a disregard for the economic contribution of wives. Easterlin's conceptualization of relative economic status focuses almost exclusively on son's income relative to father's income. Yet, a working wife, in either the family of procreation or the family of orientation, may contribute substantially to her household's income and, consequently, to the household's consumption standards. Measures of household income, couple or parental, which exclude the wife's economic contribution, then, are likely to provide poor indicators of the intended income constructs.· Turchi (1975a,b) has proposed a microeconomic model which, like Easterlin's, attempts to integrate economic and noneconomic determinants of reproduction. Like Easterlin, Turchi treats the economic variables as primary or direct determinants of demand and the noneconomic variables as secondary or indirect (through their effects on the economic variables) determinants of demand. Turchi pays particular attention to the income and price of children variables in his model. The former, he argues, should be considered as potential rather than present income, and the latter should be estimated for different sociodemographic groups. In tests of his arguments, with data from two large national surveys, Turchi found that neither his potential income nor his disaggregated price of children variables was related to the demand for children in the expected direction. Turchi attributes these failures to problems with the data, to multicollinearity, interaction effects and measurement error. It is obvious that the microeconomists have not provided convincing or compelling evidence for their arguments of rationality in reproductive

RATIONALITY AND MODELS

35

decision-making. They have not been able to demonstrate the supposed effects of their principal variables, of income, taste and/or prices, on the demand (or "production") of children. In the case of all three of these microeconomic models, the assumptions of maximization and rationality remain articles of faith, not statements of fact. The microeconomic theorists have been criticized for their excessive reliance on economic variables and for their failure to consider the effects of more subjective values and attitudes on reproductive decisions. Easterlin and Turchi are less guilty of this charge than Becker, but even they do not carefully consider the range of values and attitudes that might affect the reproductive decision. Perhaps in response to this deficiency, several socialpsychological theories of rationality in reproductive decision-making have been proposed. These theories give much greater attention to value-attitude determinants of reproductive preferences and maximization behavior. Value of Children Theory

The Value of Children theory was the first of these social psychological theories of rational or purposive reproduction (Arnold et al. 1975; Fawcett and Arnold 1973; Hoffman and Hoffman 1973). According to this theory, children are expected to provide certain positive (satisfactions) and negative (costs) values to potential parents. These perceived positive and negative values combine to form a net worth or value of children to a couple. This net value of children directly affects the couple's reproductive motivations and decisions. The higher the value of children at any given time, the more likely a couple is to have a (another) child. The value of children is seen as only one determinant of the reproductive decision. Other determinants include alternatives to achieving the values, costs of achieving the values, and barriers and facilitators to achieving the values. The value of children variable is, however, regarded as the most important determinant. The authors of the theory have developed several lists of values, i.e., satisfactions and costs, that might be perceived as important to parents and potential parents. These lists typically include such factors as self enhancement, economic utility, opportunity costs, and physical demands. The usual procedure in research is to present such lists to respondents and to ask them to rate their degree of agreement or disagreement with the values. The primary focus of research with this theory has been to compare value of children ratings with various reproductive attitudes and practices. 1 An extensive Value of Children study was conducted by Arnold and others under the auspices of the East-West Population Institute. This group

36

CROSBIE

surveyed married couples with children. Moderate size samples (180-280 couples) were drawn from each of six countries, including the U.S. (Hawaii). The survey questionnaire contained measures of 6 sociodemographic variables, 14 value-of-children variables or dimensions, and 8 reproductive knowledge, attitude and practice variables. The 8 reproductive variables included number of contraceptive methods known, general birth control attitude, situational birth control attitude, current use of birth control, number of living children, number of children wanted, ideal number of children, and additional number of children wanted. The major finding of this study is that the 14 value-of-children dimensions did predict most of the 8 reproductive variables moderately well in most countries. After controlling for sociodemographic differences, the 14 value-of-children dimensions added an average increment of 10.4 percent in explained variance over the 48 comparisons (6 countries by 8 reproductive variables). The value-of-children variables predicted the number of additional children wanted better than they predicted any other reproductive measure. The value-of-children variables did not predict current use of birth control (the only behavioral measure in the set) as well as they predicted most other reproductive measures. It should be noted, as the authors do (p. 118), that some of the coefficients for the value of children variables were in the wrong direction; thus, the explained variance attributed to the set of variables is somewhat overestimated. Bulatao and Arnold (1977) report on a restudy of three of the countries that were part of the first Value of Children study. Using a similar methodology, but with larger and more representative samples, these researchers, again, found that the value-of-children variables contributed somewhat, . though not significantly, to the explanation of additional number of children wanted. Hoffman and Manis (1979) report, from a large U.S. sample, slight positive effects of several value-of-children variables on desired family size. In neither of these studies were the value of children variables related to reproductive behavior, per se. Overall, the value-of-children variables do seem to predict some reproductive attitudes or intentions fairly well. There is little evidence, however, that these variables predict reproductive decisions and behaviors well at all. Fried and Udry (Fried and Udry 1979; Fried et al. 1980) have offered a very similar theory of reproductive decision-making. According to this theory, the higher the utility or value of consequences that couples expect from children, the more likely they are to decide to have a (another) child. Fried and Udry tested this argument in a two-year, panel study of a moderate size sample of married couples. They initially measured each husband's and wife's utilities for children and their intentions to have a (another) child. Two years later, they reinterviewed the couples to

RATIONALITY AND MODELS

37

determine their intervening reproductive behavior (attempts at pregnancy and actual pregnancy). Fried and Udry report that the utility variables explained reproductive intentions very well (R2 = .40) and reproductive behavior moderately well (R2 = .20). These results, however, clearly overestimate the explanatory value of their utility theory. Fried and Udry obtained their estimates of explained variance by regressing intent and behavior on the separate husband and wife utility measures with stepwise (forward) inclusion. This is essentially an atheoretical statistical procedure that allows for the inclusion of variables which may not be related to the explained variable in the expected or predicted manner. For example, the expectation that children will enhance one's self-esteem, which should have a positive effect on the intent to have a (another) child, is shown to have a strong negative effect and, yet, is included in the final regression equation for reproductive intention. Their reports of explained variance, then, contain not only predicted effects but inconsistent or unpredicted effects, as well. It is impossible to tell, from the data presented, just how much of the variance attributed to the utility measures is actually due to the predicted effects of these measures and how much is due to inconsistent and/or chance associations in the data. 2

Subjective Expected Utility Theory Beach, Campbell and Townes (Beach et al. 1976, 1979; Townes et al.

1976, 1977, 1980) have presented a somewhat different social psychological

model. Beach et al. have argued for a Subjective Expected Utility interpretation of the reproductive decision. This interpretation recognizes the dimension of uncertainty in decision-making. According to standard SEU theory, people assess the anticipated consequences of their decisions in terms of both value or utility (U) and certainty or subjective expectation of occurrence (SE). Any given consequence is weighted as the product of its SE and U, i.e., SEU. The decision value of any given alternative is simply the net sum of the SEU products of its consequences, and it is assumed that the decision maker will choose that alternative with the greatest or maximum sum ofSEU. The Beach theory differs from most standard SEU theories in terms of its assumption of effect. While most SEU theories assume a linear relation between magnitude of SEU and probability of choice, Beach et al. assume a step-function relation. According to the Beach theory, couples with a net positive SEU will elect to have a (another) child, and couples with a net negative SEU should decide not to have a child. There has been only one test of this theory (Beach et al. 1979; Townes

38

CROSBIE

et al. 1980). This was a two-year panel study involving 165 married, predominantly professional and well-educated couples. All couples were using contraception at the onset of the study. At the time of initial contact, each husband and wife were separately given a Birth Planning Hierarchy which provided measures for their respective subjective expectations and utilities for each of 20 parenthood consequences. Together, the husband and wife were also asked to indicate their desired family size and reproductive or birth intentions over the next two years. Two years later, the couples were recontacted and asked about their ensuing reproductive behavior. Couples who had stopped using birth control, who were currently pregnant or actively trying to become pregnant, or who had had a child were recorded as having made a positive reproductive decision. Otherwise, couples, including those with elective abortions, were recorded as having made a negative reproductive decision. Once these decisions were known, a couple's SEU score was created by regressing the reproductive decision on the separate husband and wife scores and then weighting and adding husband-wife scores in a way that would maximize predictiveness. This procedure, of course, capitalizes on chance associations in the data. The overall prediction rate for the couple SEU scores was 70 percent; that is, 70 percent of the couples predicted to make a positive or negative reproductive decision did in fact do so. This prediction rate is unusually high and may well be due to the selective nature of the sample, independent of' the soundness of the theory. The couples in the sample were generally middle class and well-educated, and all were using contraception at the onset of the study. These are people who are likely to practice control over their reproductive lives, and thus, their reproductive values and attitudes can be expe~ted to be somewhat consistent with their behavior. As evidence of this consistency, the authors report a slightly higher prediction rate (72 percent) for the couples' initial expression of desired family size and a much higher rate (88 percent) for their initial reproductive or behavioral intentions. Both of these rates are unusually high for research of this kind, and it is quite possible that the success of the SEU scores also benefited from the consistency in the sample. Thus, while the results of this research seem to support the theory, the unique characteristics of the sample bring the generality of these results into question. Theory of Reasoned Action Davidson and Jaccard (1975, 1976, 1979; Jaccard and Davidson 1972, 1976; Fishbein et al. 1980) have used the Theory of Reasoned Action to explain both contraceptive and reproductive decision-making. According to

RATIONALITY AND MODELS

39

this theory, a person's behavioral and normative beliefs about some behavior jointly determine the person's intention to perform the behavior; this behavioral intention, in turn, serves as the sole and direct determinant of the behavior itself. The more positive the behavioral and normative beliefs, the greater the intention, and the more likely the behavior is to be performed. The behavioral and normative beliefs serve as the major explanatory variables in this theory. As Jaccard and Davidson have noted (1976:330), the conceptualization of behavioral belief is essentially identical to the conceptualization of subjective expected utility. The behavioral belief variable is treated as the sum of the products of the beliefs (B) or subjective expectations that certain consequences will occur from the performance of a given behavior, and the evaluations (E) or utilities of those consequences if they do occur. The normative belief variable, in contrast, is treated as the person's perception of significant others' evaluations of the behavior and the person's motivation to comply with these others. The Theory of Reasoned Action is not a theory of strict rationality, since it assumes that the tendency to maximize satisfaction, or to choose the behavior with the greatest sum of BE, will be moderated by the normative beliefs concerning the behavior. If a person's behavioral (BE) and normative beliefs are congruent, then the person's intentions and behaviors will tend to be consistent with behavioral beliefs and the person will appear to be rational. If on the other hand, the person's behavioral and normative beliefs are incongruent, then the person's intentions and behaviors will tend not to be consistent with behavioral beliefs and the person will appear not to be rational. Since our concern, at this point, is solely with rationality, and not with factors that may moderate rationality, we, shall limit our consideration to the behavioral belief (BE) component of the theory. Davidson and Jaccard have argued that there are consequences of both contraception and reproduction which affect persons' behavioral beliefs and, subsequently, influence their contraceptive and reproductive intentions and behaviors. Davidson and Jaccard have, thus far, conducted two tests of this argument. In the first of these tests, with a small sample of female college students, Davidson and Jaccard (1972) measured the beliefs (B) and evaluations (E) of 15 potential consequences of using birth control pills. They also measured the womens' intention (BI) to use the pill. They report that women intending to use the pill did differ significantly on some beliefs and some evaluations from women not intending to use the pill. They provide no information, however, on the degree of association between B and E and BI or between BE and Bl. Nor did they provide any information on actual usage or behavior.

40

CROSBIE

The second test of the theory involved a moderate size sample of married women. This was a two-year panel study in which the researchers initially measured the beliefs and evaluations of 9 potential consequences of having a (another) child within the next two years (BE) and the behavioral intentions (BI) of doing so. Two years later, the women were reinterviewed to determine whether they had or had not attempted to have a (another) child (Beh). In the first report of their results, Davidson and Jaccard (1975) found a strong positive correlation (r = .74) between BE and BI; that is, the behavioral belief of the consequences of having a child did predict the intention to have a child quite well. In a follow-up report, Davidson and Jaccard (1979) indicate that the correlation between BI and subsequent Beh was also quite high (r = .62). Unfortunately, the researchers did not report the correlation between BE and Beh. However, if we assume their underlying argument (BE ..... BI ..... Beh), we can estimate this correlation to be r = .46. This would indicate a rather substantial predictiveness for BE, r2 = .21. Thus, it would seem that Davidson and Jaccard's respondents were somewhat consistent in their reproductive decisions. Smetana and Adler (1979) applied the Theory of Reasoned Action to the decision to have an abortion. These researchers administered a questionnaire to a small sample of women who were receiving pregnancy tests. The questionnaire contained, among other items, measures of the beliefs and evaluations of 53 potential consequences of having an abortion and a measure of the intent to have an abortion if pregnant. Respondents found to be pregnant were to contact the researchers at a later date to inform them of their abortion decision. The results of the research are presented in path analytic form without all of the necessary correlations. It is possible to estimate th~ correlations from the path model, however, and these estimates suggest that the correlation between BE and BI was .29, between BI and Beh was .96, and between BE and Beh was .27. This would indicate an insubstantial explained variance of r2 = .07 for the behavioral belief variable. Thus, while the women in this sample certainly followed their intentions, neither their intentions nor their behavior was strongly predicted by their behavioral beliefs. This review of the major models and tests of models of rationality in reproductive decision-making indicates that the empirical support for these models and their underlying assumptions is weak or non-existent. The support is certainly not compelling. While there is some evidence that reproductive preferences, as measured by economic and value and attitude indicators, do seem to influence stated desires and intentions, there is little consistent evidence that these preferences exert much influence on subsequent behavior. People may have preferences for the consequences of their reproductive behavior, but these preferences seem not to be sufficient to

RATIONALITY AND MODELS

41

determine their decisions and behavior. Criticisms of Rationality Theories These conclusions should not surprise the many students of reproductive behavior who have criticized the rationality models. These persons have argued that the reproductive decision process is likely to be much more complex than has been portrayed by the rationality theorists; that while people may hold a priori preferences for their behavior, these preferences are not likely to solely or even largely determine their behavior. Several critics of the rationality models have suggested that reproductive preferences may influence reproductive decision-making only for certain kinds of people or only under certain conditions. Smith (1978), for example, who is suspicious of the reductionist bias in rationality theory, has suggested, nevertheless, that persons with a high sense of efficacy or internal locus of control may be more likely to follow their preferences in reproductive decision-making than persons with a low self or an external locus of control. Leibenstein (1974) has argued that the calculation of consequences may only occur for people contemplating marginal births or births that may occur after desired family size has been achieved. Other critics of the rationality models have suggested that there are competing variables or preference factors which may interrupt or moderate the influence of reproductive preferences. Blake (1968), Dusenberry (1960), and Ryder (1973) have argued that norms of family formation and family size may exert an important influence on reproductive decisions. For example, Catholics tend to want and to have more children than non-Catholics, independent of income. Similarly, it is less appropriate for nonmarried women to conceive than it is for married women. (Davidson and Jaccard do give normative expectations an important moderating ·role in their full model, but theirs is a model of reasoned, not strictly rational action.) Bean (1975) and Namboodiri (1975) have critized the microeconomists for failing to differentiate between husband and wife preference orderings. The microeconomists assume that these different orderings will be harmoniously blended into a unitary family or household preference ordering. While this blending may occur in some families, it is not likely to occur in all couple relations; it is particularly not likely to occur in nonmarital relations where the couple may have little commitment for conflict resolution. In these instances, and perhaps in all instances, it may be more accurate to consider the male or husband's preferences as separate from and supplemental to the female or wife's preferences. Fried and Udry, Beach et al., and Davidson and Jaccard have treated husband and wife preferences

42

CROSBIE

separately, and all have found that husband's preferences do provide a supplemental source of influence on reproductive decision-making. Leibenstein (1981), who has been a persistent critic of the rationality models, has suggested that habit may play an important supplemental role in reproductive decision-making. Leibenstein argues that as day-to-day behavior becomes routinized, for whatever reasons, it is often easier to follow the routine than to change it. The routine solves the problem for which the decision must be made and, thus, provides an alternative basis for decision-making. Finally, Namboodiri (1975) and Perlman (1975) have criticized the microeconomists for failing to incorporate the effects of uncertainty in their models. The uncertainty of the future (e.g., of income or wife's employability) seems very likely to affect economic-based decisions (see Hayden, this volume). Beach et al., and Davidson and Jaccard do incorporate uncertainty (subjective expectations, beliefs) in their models, as a means of weighting the consequences of parenthood, but they do not include all relevant uncertainties in the situation. For example, the certainty-uncertainty of conception without contraception seems very likely to effect the decision to use contraception, independent of the weighted benefits and costs of parenthood. These arguments do not exhaust the criticism that has been made of the rationality models of reproductive decision-making. They do, however, reflect the general tone of the criticism. Critics of the rationality models feel .that people do not or do not always make their reproductive decisions on the basis of a simple preference ordering of expected consequences; they feel that there are additional conditions or contexts which must be taken into account before we can fully predict reproductive decisions and behavior. To date, these critics' arguments seem not to have been heard by the proponents of the rationality models. Although several of the rationality theorists have tried to incorporate other variables into their models, they generally treat these variables as indirect rather than as direct determinants of decisions and behavior. With the exception of Davidson and Jaccard, who regard normative beliefs as codeterminate with behavioral beliefs, rationality theorists continue to treat the preference ordering of consequences as the sole determinant of decisions and behavior. This lack of attention to these criticisms may be due to the fact that, like the rationality models themselves, counterarguments have not been accompanied by an empirical demonstration of their validity. This paper provides such a demonstration by simultaneously evaluating the expanatory power of rationality variables and those suggested by critics of the rationality approach.

RATIONALITY AND MODELS

43

The Data The data reported here were collected in conjunction with a comprehensive test of Luker's rationality theory of contraceptive decision-making (Crosbie and Bitte 1982). The data were gathered with a survey questionnaire and included measures relevant to the test of Luker's theory, including measures of sociodemographic variables, measures of the expected benefits and costs of contraception, pregnancy and parenthood, and measures of contraceptive use and the frequency of contraceptive risk-taking. The survey questionnaire also contained measures of variables directly or indirectly implicated in the arguments of other rationality theorists and in the arguments of the critics of the rationality explanations. And the questionnaire contained measures of some reproductive intentions and reported reproductive behaviors. Thus, the data should provide a comparative look at the rationality and the alternatives to rationality arguments presented here. The questionnaire was administered to a nonrandom sample of 567 male and female junior college and university students. The sample included 221 currently sexually active females, and the results to be reported will focus on this subsample (although, the results are nearly duplicated for a similar subsample of males). Of the 221 respondents in the sample, 43 were married and 178 were unmarried. Ninety-eight of them had had one or more pregnancies. Their median age was 21.6 years; their median level of education was 14 years; and their median sexual involvement with their most recent partner was between 10 months and one year. The women had a median of 8.5 acts of intercourse in the past four weeks; in this time, 168 had taken no contraceptive risks and 53 had taken one or more risks.

Reproductive Variables The concept of rational or purposeful reproduction entails a chain of distinct decisions. Persons who are determined to have a high preference for children must first decide to have a (another) child, must then decide to conceive, and must then decide not to use contraception to prevent conception. Conversely, persons who are determined to have a low preference for children must decide not to have a (another) child, must decide not to conceive, and then must decide to use contraception and to use contraception consistently. Thus, if people are rational or purposeful in their reproductive decision-making, we would expect the determinant(s) of their preference for children to be positively related to their intention to have a (another) child,

44

CROSBIE

positively related to the attempt to conceive, and negatively related to the use and to the consistent use of contraception. A measure of each of these reproductive decision variables was included in the survey questionnaire. The respondents were asked how many children they presently had and how many children they ultimately planned to have. The difference between these responses indicates the number of additional children planned. This difference was dummy coded to create a DEMAND variable, 1 if one or more additional children were planned and 0 if no additional children were planned. The respondents were also asked if they were currently trying to becom~ pregnant or to conceive. Responses to this question were dummy coded to form a TRYPREG variable, 1 if currently trying to conceive and 0 if not. The respondents were asked to indicate their frequency of sexual intercourse over the preceding four weeks and the number of times in this period that they had not used any form of contraception. The responses to these questions were used to construct three contraceptive variables: ANYUSE, 1 if they had used some form of contraception at anytime in the preceding four weeks and 0 if not; CONSIS, 1 if they had used contraception everytime and 0 otherwise; and FRQEXPOS, the frequency of not using any contraception or exposure to pregnancy risk. TABLE 1

CORRELATIONS AMONG REPRODUCfiVE VARIABLES

Demand Trypreg Anyuse Consis

Trypreg

Anyuse

Consls

Freqexpos

.08

.02 -.07

-.16* -.16*

.01 .29*

.40•

-.24• -.59*

*P .40). These results suggest that there is no effective linkage between the perceived value of children and the fertility pattern in the El Turnbo population. I infer that effective fertility decisions have not been made and do not account for variance in fertility in this population. If fertility decisions do not account for variance in fertility within this population, we may infer that fertility is due to behavioral contraints on the frequency and timing of coitus and post-partum infecundability, which may vary in different economic niches (Handwerker 198la, 1983). To examine this possibility I regressed on the number of live births: women's work load (an ordinal variable measuring intensity of work load), male absenteeism (an ordinal variable measuring the frequency of the husband's absence from the household due to wage labor or commercial activities), length of lactation, infant mortality, age of marriage, years out of a sexual union, and the proportion of time a woman has spent in her reproductive period (age 15-49). All variables except infant mortality and proportion of reproductive span varied in only minor ways over time and across economic strategies

152

7()..74 65-69 6()..64

5~59

5()..54

45-49 4()..44 3~40

3()..34 2~29

2()..24 1~19

10·14 ~9

0·4 35

30

25

20

15

10

5

0

5

10

15

20

FIGURE 1. AGE/SEX PYRAMID. EL TUMBO, 1978.

25

30

35

40

«
20 I I i I

~~~;~;~ ~;~;1

::::::::

mt~

tt Jt

;~~I~~ ::;:;:;:: tt=

0

nnm

= never married (N,C)

35-39 40

·::::::::

~~~~ II

·:·:·:·:

::::::::

1~~~~; ··:·:·:·:

········ ~~~;~;~~;

:::::::

:·:·:·:

;;;;~~;

rm

·!!!!!!!

:~~I~ .:::::::

rmm

-

30-34 34

:f~:;

;. ;

~~~rt

I~l ·:::::::·

.:::::::·

;!1!1!1!!

I~~I~ ::::::::

IJ.LLLIIwmm

Marital status of females by age groups. Zorse, household survey.

N = never married C = poaeling M = never married

0-4 94

:;:;:~:~

l!l!l!l!

~~;~;~~~~

u

III

I

!I~~~~

.·:·:·:·:

::::::::

;;:::::::

N

00

N

THE POLITICAL ECONOMY

283

The difference in the average age of marriage of men and women combined with the age structure of a rapidly growing population {Figure 2) accomodated the practice of polygyny. In December 1977, 35 percent of the men in the Zorse sample had more than one wife. Most such men, however, had no more than two wives. As elsewhere in Africa, ''marriage'' in Zorse is difficult to define precisely. Unlike birth dates, I did not try to improve the precision of the marriage date estimates reported. People in Zorse say that women are marrying at younger ages than they did in the past.·Although their account cannot be tested against precisely defined ages at marriage, it is consistent with an observed reduction in the age at first birth. Thus, whereas prior to 1948, 66 percent of first births occurred to women under 20 years of age, between 1948 and 1977, 79 percent of first births occurred to women under 20 years of age. The mean age at first birth for women born between 1908 and 1957 fell1.5 years from 19.5to 18.0.1

Decreasing Postpartum Infeeundabllity The increasing value of the Index of Postpartum Infecundability {Ci) reflects the decreasing effect of lactation and abstinence. The period of postpartum infecundability includes the minimum post-natal infecundability, lactational infecundability, and voluntary abstinence. Together with waiting time to conception, time added by intrauterine mortality, and gestation, they make up the birth interval. Ci is estimated as: 20/18.5 + i (Bongaarts 1978:115-116). The numerator is the average estimated birth interval without lactation or postpartum abstinence. The denominator is the 20 months minus 1.5 months of infecund period following birth, and i is the total infecund period, i.e., 1.5 plus the effect of lactation and postpartum abstinence. Thus, Ci estimates the increase in the birth interval added by lactation and abstinence. In 1963-77, the mean birth interval in Zorse was 38.6 months, so that i = 20.1 months, and Ci = 20/38.6 = .518. In 1948-62, the mean birth interval was 39.2 months, and Ci= .510. In 1928-47, the mean birth interval was 42.9 months and Ci = .466. As seen in Figure 7, postpartum infecundability reduces TF by 50 percent, and is much more important than the increasing proportion of married women. As elsewhere in Africa {Caldwell1977b, Retel-Laurentin and Benoit 1976; Dorjahn, this volume; contrast Weil, this volume) of the two components of Ci, lactational infecundability is of negligible importance. Breastfeeding is universal in rural northeastern Ghana and Kusasi women breastfeed their infants freely, and sleep with them at night.

284

CLEVELAND

Nonetheless, lactational amenorrhea is consistently shorter than the period of breastfeeding. For instance, Mondot-Bernard's (1977:23) review of seven studies of lactational amenorrhea in Africa revealed a range of .76 to 1.56 years duration of amenorrhea. In all but one study breastfeeding lasted from .19 to .68 years long than the amenorrhea. In Zorse, 86 percent of all birth intervals are two years or longer, and probably are not greatly influenced by lactational amenorrhea. Sexual abstinence is not only the most important determinant of postpartum infecundability, it has long been consciously used by Kusasi to regulate their fertility and, so maximize the number of surviving children. As explained in the previous section, the personal insecurity caused by local feuding and the initial impact of colonialism placed a premium on birth intervals long enough so that the next to youngest was fully capable of running away on his own. Men had to carry bow and arrows and were not expected to help with the children. More important, however, people understand that if the woman becomes pregnant before the youngest child is mature and healthy enough, breast milk will dry up, the child will not receive as much of his mother's attention, and the child's health will be in even greater jeopardy. The unfortunate child whose younger sibling is born too early is recognized by his big belly, leaness, and weeping. The length of abstinence is not formally specified. The decision to resume sexual relations is a joint one and depends primarily on the health of the youngest child and restraint of the married couple. A Kusasi child should be able to walk and talk before the parents resume intercourse, and this generally occurs by age 1.5 to 2. In addition, however, a child also must be healthy enough to take care of himself. This emphasis on the health of the child is not new in Kusasi culture. Rattray's Kusasi informants of 50 years ago pointed out that a man does not have intercourse with his wife for two or three or more years postpartum: "it depends on how soon the child is walking about"(l932:387), and walking may be delayed because of chronic poor health. Indeed, this conscious use of birth spacing to maximize the number of surviving children may be widespread in Africa. The nearby Tallensi, among whom nursing mothers do not have sexual intercourse, have views very similar to those of the Kusasi. Fortes (1949:20) states that it is "not a ritual matter, subject to mystical sanctions, but a practical necessity in order to prevent the women from conceiving again too soon.'' Coitus interruptus is sometimes practiced by the Tallensi to this end. RetelLaurentin and Benoit came to equivalent conclusions in their study of the Bobo-Oule of Upper Volta (1976). They criticize other studies of African populations for assuming that there is a uniform period of abstinence until the child is walking. The official criterion is "that the infant is toddling or that the mother has ceased lactating. . . .If the infant is not doing well,

THE POLITICAL ECONOMY

285

however, breast feeding may be continued until the danger to its life is past" (1976:291 ,292). If the Kusasi husband tries to have intercourse against his wife's wishes, she is justified in refusing him and will tell his mother and father who will reprimand him if they agree with their daughter-in-law. One informant said that in the old days if a young man's wife had children too close together, his household head would beat him. Some men have sexual access to other women, either their other wives or a single woman. In this case, it may be the woman who tries to force herself on her husband before the youngest child is old enough. In one group interview with a husband and his two wives, all agreed that a woman may take the initiative to resume intercourse because she feels sexually deprived. Increasing Supply of Children Because postpartum abstinence is a socially controlled behavior that to a great extent is consciously adjusted to maximize the number of surviving children, it has changed through time as factors affecting survival have changed. Improvements in nutrition, public health, and medical care in the West African savanna, especially since World War II, have been the major cause of declining infant and child mortality (Figure 9) and increased longevity (Gaisie 1976: 137-39). In Kusaok, Western medicine is increasingly popular for treating children because children have the highest mortality rates and have not become completely integrated into the social system. In 1974-75, for instance, out of an eligible population of about 50-60,000 children in the eastern (Agolle) portion of Bawku District, there were 23,000 new attendees at child welfare clinics (BDMHS 1975). In Zorse, over 50 percent of the children under 8 had attended the clinic at least once, and many had been vaccinated against polio, measles, diptheria, and tetanus. Out of 148 illnesses occurring during · the previous two weeks reported for 579 Zorse children in April and July, only 20 percent received no treatment. Traditional Kusasi medicine was obtained for 13 percent, Western medicine for 61 percent, and 7 percent utilized both traditional and Western medicine. Not only do children become ill less frequently in Kusaok, many people have taken the view that if a child does become sick he can be cured more easily now than in the past. Consequently, some people now believe that the health of the child is no longer particularly important in making the decision to resume coitus postpartum. Under increasing resource stress, Kusasi parents generally have taken advantage of declining child mortality to increase fertility by shortening the period of postpartum abstinence.

0

Figure 9.

15 69 .217

11 65 .169

8 24 .333

3 12 .250

I

19431947

.

1938· 1942

.

1933· 1937

I

1928· 1932

0

.15

.20

26 98 .265

I

19481952

29 109 266

I

19531957

24 134 .179

I

1958· 1962

37 165 .224

.

19631967

39 183 .213

I

1968· 1972

Proportion of live born children dying before age five, by year of birth. Zorse, pregnancy history survey.

5q0

Numerator Denominator

Year of birth

1.0

C'

.25

.30

.35

IV 00 0\

THE POLITICAL ECONOMY

287

Figure 10 demonstrates that most of the deaths associated with short birth intervals occur before the conception of the next child. Below the death-age= birth-to-conception interval the correlation between birth interval and death age is .72 (P

::Jo

oE

U::!!?

2.o

2.5

3.0

3.5

Figure 11.

I

I

I 19081912

14

I 19031907

14

41

I 19131917

r = 0.21, p < .001 F = 3.93, p < .001

77

.

19181922 19281932 109

50

.

19231927

median

19381942 145

77

91

19431947

= 3.20. n =

19331937

i for 1903-1957

+

61

19481952

702

23

19531957

x and standard deviation

Birth intervals to the next youngest for mother's birth year, grouped in five-year intervals, 1903-07 to 1953-58. Includes only children alive when next youngest conceived, and only birth intervals less than 5 years. Zorse, pregnancy history survey.

n

Mother's birth year

iii

=

·=....

~

c:

a;

.2::

as Q)

~

4.0

4.5

N 00 \0

1/7 8/27 4/26

' ',',,

1117 11/64 12/69

I 2.5-3.0

' ''

8/45 11/92 111105

I 3.0-3.5

2/18 2/30 2/33

I 3.5-4.0

1/31 0/18

0127

4.0-4.5

0/10

017

0/9

4.5-5.0

- - - - 1928·1977

1928-1947

••••••••••••••••• 1948-1962

-----1963-1977

Figure 12. Mortality rates to five years by birth interval for those alive when next youngest conceived. 5qX is the mortality rate of the first born from the time second is born (x) to 5 years. Zorse, pregnancy history survey.

0 0.7·2.5

1928-47 1948-62 1963-74

''

1

''

Birth interval

.100

.....

X CT

.200

.300

~

THE POLITICAL ECONOMY

291

1978:96-97). In the Kusasi case, however, children do not necessarily consume more than they produce, and increased survivorship need not shift households from a situation of excess demand to one of excess supply. That shift is a function of the relative costs and benefits of births and children, and those costs and benefits are relative to specific productive conditions. Figure 13 indicates the relationship between the number of live births (B), and the number of surviving children (Cs) and parental consumption (Op), when children produce more resources than they consume. The first figure (1aa) shows that there is a threshold level beyond which the number of surviving children decreases as the number of births increase. This threshold is the result of increasing mortality of young children born at very short intervals in societies where their survival during the critical first 2-3 years depends on their mothers' constant attention. The second figure (13b) shows that past this threshold, the goods and services parents consume decreases as the number of births increase. This threshold appears to vary with the available health care and with production technologies, i.e., with the supply of births, survivorship, and the demand for births and the labor of children. As intensity of agriculture increases, returns to labor often decrease as more labor is invested in each unit of land (Boserup 1965). Under these conditions, children become labor resources for the household at an early age, alleviating the strain on their parents. In labor-intensive farming communities throughout the world (see Cain 1977 for Bangladesh; Ho 1979 for the Philippines; Nag, White, and Peet 1978 for Java and Nepal; and Caldwell 1976, 1977a,b, 1978, for broader analyses), children appear to be producers of wealth, not consumer durab1es. Where fertility controls exist in these communities, they plausibly are used, as they are among the Kusasi, to maximize the number of surviving children, and thereby to maximize the resources available to parents. Where, as among Kusasi, resource stress increases, it is not reasonable to suppose that an increasing supply of births and/or children will alter reproductive strategies. Demand for births can decrease only with a change in the labor value of children and a reduction of resource stress (see Hayden, this volume). For the Kusasi, a change in farm technologies that increases the productivity of labor, and a change in social organization that links households' short-term demand for labor with long-term management of community agricultural resources are the most plausible stimuli for a fertility transition. Notes

Acknowledgements. I thank the people of Ghana, Bawku District, and especially the chief, elders, and all the residents of Zorse, for their support of my work with them.

292

n

0~~--~------~ n 8 0 g

0 ....._ __:.._ __

0 Figure 13.

8

n

Relationship between number of live births and key variables in the cost benefit model. B = live births, s = survival rate, .5 for Keneba, P = .38 for Manduar). The continuous increase in fertility in Keneba, however, might indicate a real shift in birth rates, and a difference of proportions test comparing fertility at the beginning and end of this period suggests that in Keneba fertility has increased over time (P = .03). The apparent downturn in mortality for Keneba could be found by chance 7 times out of 100. The mortality differences recorded for Manduar plausibly are real (P =.013), but these data suggest that mortality has increased (by 141 OJo) between 1951 to 1975. I infer that the basis for population growth in these villages has been a reasonably constant disparity of births over deaths, and a slight increase in fertility in Keneba.

Fertility in Keneba and Manduar With the exception of culturally-prescribed periods of infecundability, throughout the 25 year study period women were continually exposed to conception during their reproductive lives (Billewicz and McGregor 1981:234; Caldwell and Thompson 1975:524). Seventy percent of the observed women had their first child before age 20, and 36 percent had their first child before age 18. Seasonality in conception is not noticeable except for a birth deficit in the May-July quarter suggesting a lower rate of conception in the August-October quarter. This quarter is the one in which women are most intensively engaged in rice production, and sustain their largest

AGRICULTURAL INTENSIFICATION

309

Calorie deficit (Haswell 1981:39-40). Billewicz and McGregor (1981:234) suggest that this decrease in conception may be related to a high incidence of lactation in the wet season, and Lunn et al. (1980:234) document an increase in prolactin levels during this period due to nutritional deprivation (also see Harrell 1980:798; for a review, see Wilmsen, this volume). Due both to migration of men and heavier labor requirements for women, coital frequency may also fall off (see Odell, this volume, and 1982). As one would expect, birth intervals are a function of post-partum infecundability. The length of this period varies with infant mortality and the length of post-partum lactation (Billewicz and McGregor 1981:234-5; cf. Bongaarts 1982). The MRC study does not present data on variation in post-partum lactation over the study period. Data from other sources suggest that lactation periods have declined over this period (see below). Figure 2 exhibits the age-specific fertility schedules for Keneba and Manduar, and Figure 3 compares the age-pattern of childbearing in these villages with those characteristic of natural and target fertility populations. These comparisons make clear the parity-independence of births in Keneba and Manduar. Total fertility rates for these populations do not appear to be explained by the use of contraceptives or induced abortions. In Keneba and Manduar, the abortion rate is reported to be "very low" and antinatal practices are little known (Caldwell and Thompson 1975:499). Research in Bumari and Bantunding are consistent with MRC findings and suggests that in these villages abortions were induced at most once a year. Instead, total fertility appears to be a function of the frequency of coitus and the length of postpartum infecundability (see Handwerker 1983). Since 1950, variables bearing on the frequency of coitus have not changed dramatically. Both divorce and widowhood, for instance, are associated with a prescribed celibacy period of 4 months, and divorce rates are high, particularly among women aged 25-34 (Caldwell and Thompson 1975:522; Billewicz and McGregor 1981:220; Weil1976). But there is no evidence suggesting an increase in the rate of either variable over this time period. Celibacy periods have become attenuated, but in any event, it is unlikely that the prescribed period of celibacy affects total fertility appreciably because of increasing rates of remarriage. The labor potential of women makes them strongly attractive as mates. Bridewealth paid for women re-marrying between the ages of 25-34 tend to be higher than that paid for their first marriage. Such women rarely wait four months to move to their new husband's households if they are not pregnant. The mean would be closer to 6 weeks. in addition, despite the prescribed celibacy period for widows, Mandinka rules include a strong preference for leviratic marriages of a widow to a brother of the same compound as the deceased husband. If the woman is of reproductive age and

310

.40

.35 .30

I

~.25

§

r·20

'., %

... .15

a: ii

20

25

30

35

AGE CLASS

40

45

50

Fig. 2. Age-specific fertility in Keneba and Manduar, The Gambia(1951-75). F'lgUre based upon data in Billewicz and McGregor 1981 : 237 (Table 14, Method "8").

311

_ _ Coole-Trussell Parameters Kenebo Monduor Xfor Tor9et Fertility Populations

1.10 1.00 0.90

XN

11.111.1

Sl Sl

.80

!;(!;(

\

. \

\

(/) (/) .70 0::0::

' I zz

.\

\

.60

0 0.50 (/)(/) 0::0:: 11.111.1

0.. 0.. 40 ............. (/)(/)

t: t: .30

:I: ::I:

a; a;

.20

.10 .00~----~----_.----~~----~----~-=~~

20

25

30

35 40 AGE CLASS

45

50

Fig. 3 Age pattern of childbearing in Keneba and Manduar, The Gambia (1951-75). Figure based upon data calculated from Billewicz and McGregor 1981: 237 (Table 14, Method B).

312

WElL

has not been socially-defined as infertile, any man who can benefit from the leviratic rule tries-usually successfully-to marry her. In such marriages, women tend to remarry about 40 days after the funeral. This pattern appears to have begun at Bumari in the 1950s and at Bantunding in the 1960s. There is mixed evidence that fertility may be lowered in polygynous marriages (e.g., Dorjahn 1958; Isaac 1980; cf. Handwerker, this volume) and in this strongly patrilineal and patrilocal society polygyny is a preferred marriage form. The incidence of polygyny.does not appear to have changed significantly through the 1951-75 period, remaining at a level of about 40 percent of all families. 5 The one variable affecting the frequency of coitus that may have changed significantly over this period is the average length of spousal separations due to migratory wage labor. No data exist with which to evaluate this issue, but it is possible that the increasing incidence of male migration has meant lower coital frequencies and a decline in fertility. However, even if migratory spousal separations have implied a lower fertility, declines in the period of post-partum infecundability appear to have more than compensated. Traditional Mandinka rules acknowledge the adverse health effects associated with short birth intervals and stipulate that postpartum breastfeeding should be continued for two years, and that coitus is proscribed over this period and for an additional 9 months after weaning. It is reasonable to presume that these rules-if not followed precisely-were at least approximated in the past. At least since the 1960s, however, neither rule appears to have been followed closely. Although all children were breastfed for at least 18 months in an MRC study in the early 1960s, only 33 percent of the sample children were breastfed a full two years. 6 Although couples may avoid coitus throughout periods of short-term breastfeeding, MRC results show that among the long-term breastfeeders only about half the couples avoided coitus throughout the period of lactation. The proscription on coitus tended to be broken at about 8-9 months. We can expect a decline in breastfeeding from an average of 24 to 18 months to reduce the period of postpartum amenorrhea from an average of about 17 months to just under 12 months (Bongaarts 1982: 188). If coitus has already resumed, fertility can be expected to increase markedly, possibly by as much as 20 percent. In the present case, the regional-wide average probably has not dropped a full six months on average. However, considering the structural changes that have occurred in rural household-production units over the past 30 years and the increasing pressure on women to be pregnant, almost certainly the average period of postpartum amenorrhea has dropped, the average length of the post-partum coital avoidance has declined, and fertility has risen.

AGRICULTURAL INTENSIFICATION

313

"Development" and Fertility in the Gambia

The continuously high, even slightly increasing, fertility documented by the MRC study is an effect of processes of economic "development" which have led to underdevelopment. Agricultural commercialization over the period 1951 to 1975 has led to demonstrable increases in rural per capita income. Between 1966 and 1976, for instance, rural per capita income rose by lRO percent, from $25 to $70. However, this change merely reflected greater than a 200 percent increase in the absolute farmgate prices for the principal cash crop, peanuts (see Dunsmore et al. 1976). Increases in real income have been negligible, for the increased cash earned by farmers has been used either to purchase food or to pay the interest on food purchased on credit. The overall trends in morbidity and mortality may not have increased over the past thirty years, but only because, in the face of a substantial weakening of local economic and social support and assistance in time of need, and a weakening of the capacity to produce due to the deterioration of rainfed-land resources, women are working significantly harder than they did in 1951. Women are working harder because, as men shifted their work activities to commercialized agriculture and migratory wage labor, women's food and cash-crop production activities have shifted to increasingly intensive forms of land use. In the absence of technological change, intensification of agriculture has been accompanied by decreasing returns to labor. The "development" processes experienced by Gambian farmers sustain a standard of living equivalent to or worse than that of the nineteenth century. As judged by the access to food through production or social mechanisms, Gambian farmers' standard of living has markedly declined. High fertility is being generated by this deteriorating standard of living. Notes Acknowledgements. I thank the National Science Foundation and the University of Delaware Research Foundation for supporting my research in the Gambia in 1966-67 and 1975-76 respectively. I am indebted to the University of Delaware Research Foundation for their generous support for the preparation of the four figures included in this paper. The views expressed in this paper do not necessarily reflect the position of either institution. I also greatly appreciate the many editorial suggestions of Paul Alexander, Mary E. Granica, Penn Handwerker, Judy Villamarin, and Juan Villamarin. 1Linares

(1981) describes a process analogous to that presented here and below through which the Diola of the Lower Casamance (south of The Gambia) are

314

WElL

shifting male labor into upland peanut and cassava cash crop production and increasing women's participation in staple rice production (also see Weil 1981). 2All techniques for which data are presented are totally manual and involve little

or no use of chemical fertilizers. However, peanut fields were occasionally browsed during the dry season by Fulani-owned cattle which dropped some manure on the fields. Moreover, women used household garbage refuse and manure from goats and sheep on their swamp rice seed beds. Producer prices were not available for the 1947-50 period. The prices for the 1973-74 year have been selected. These prices are used to create a common price demoninator for the kilogram weights of peanuts and rice. Their use for comparative purposes is justified by the constancy of their relationship to one another over the years (cf. Haswell 1975). Prices were taken from Dunsmore et al. (1976:326,380). Haswell's production cost figures are gross, and no production costs are provided for the 1947-50 period. Costs in kilograms are calculated here from UNDP 1980-11-49, and are within the range of research data from other areas of Kiang and Wuli. I computed a composite replacement value for a kilogram of rice on the following assumptions: (1) the rice produced during the three economic years met, on average, about two-thirds of the rice needs of the village, and (2) about one-third of the remaining rice needs were purchased on credit at an interest rate of approximately 100 percent for periods of 4-8 months. Haswell notes that for the period under consideration, rice constituted by weight 80 percent of the total grain needs of the community (1975:42). Rice produced within the community constituted 71.1 percent of the total weight of grains produced (2101 kg. per ha. of the total 2916 kg. per ha.; Haswell 1976:46). The food grain purchased was virtually exclusively rice, and for rice to constitute approximately 80 percent of the food grains consumed would require the purchase of 1050.5 kg. This amount would mean that one-third of all rice consumed was purchased. Drawing from retrospective data from the Kiangs gathered in the mid 1960s and estimating from Haswell's data on purchases and credit for the 1947-50 period, an approximate figure of one-third of the purchased rice being purchased at 100 percent interest is derived (Haswell 1975:55-70). This means that approximately eleven-twelfths of the rice was valued at the producer price (here D.243) and one-twelfth was valued at a price including 100 percent interest of D.486. The resulting composite replacement value was D.263. The use of production figures per hectare in Table 1 is essential for comparative purposes. However, it should be noted that a female worker rarely produced on a total area of either type of rice production that was a full hectare in size. The figures that follow are for production year 1949-50 and are treated as representative of the entire period 1947-50 (Haswelll975:22). The mean total area for a female worker in swamp rice production was .51 ha. (.06 ha. for seed beds and .45 ha. for transplanted plots) and the total area per female worker for upland rice production was .16 ha. Moreover, the total number of hours per female worker was correspondingly lower per economic year than that indicated in the table. Haswell (1975:50-51) indicates that women and girls working both types of rice production each worked an average of 155 days of 6.4 hours for a totil of 992 hours each. Of those hours, a

AGRICULTURAL INTENSIFICATION

315

female worker devoted 634 hours to swamp rice production and 358 hours to upland production (Haswell 1966:109). In sum, female workers expended a total of 955 hours on .67 ha. in both types of rice production together. It should also be noted that these data for swamp rice production do not include the hours devoted to walking to and from transplanted plots. Transplanted plots are commonly one to three miles from home. The walking often is arduous because much of the distance is through muck and even more arduous during the transplanting season when a worker must carry bowls of seedlings weighing 23 kilograms or more. Yield weights in Table 1 are for rice paddy, threshed but unhulled rice. The mean yields per female worker were 585 kg. on .45 ha.of swamp rice and 112 kg. on .16 ha. of upland rice. Male workers did not work a full hectare in peanut production the number of hours cited in Table 1. Males worked an average of 110 labor days of 5.4 hours each in all production activities, including those in peanut and millet production, resulting in a total of 594labor hours (Haswell1975:50-51). Of the total days, 71.5, or 386 hours, were devoted to peanut production (Haswell 1963:108). These hours constitute 59.4 percent of the hours devoted to one hectare, which as a proportion of one hectare, indicates the average area per male worker to have been .594 ha. Moreover, the yield weights in the table are for unshelled peanuts. The conversion rates to wheat equivalents presented. in Table 1 were taken from Clark and Haswell (1970:102). 3When poor farmers made use of ox-plows, predominantly either they hired the ox-plow teams of wealthy neighbors or they purchased cheaper, used plows from Senegal and rented animals. In both cases, however, they cultivated less land than wealthy farmers and their yields per land unit were lower (Weil1970, 1978). 40f course, nuclear families appear as phases in the developmental cycle of domestic groups even where extended family production units are institutionalized. In The Gambia, however, the increasing prevalence of nuclear family production units reflects a change in family and agricultural institutions. 5However, it is possible that the incidence of polygyny may be declining, or may shortly in the future. Polygynous unions are most common in extended family households. For instance, at Bantunding in 1976, 35.3 percent ofthe nuclear families were polygynous whereas 71.4 percent of the extended families included at least one polygynous union. The shift to nuclear family production units thus may be accompanied by a declining incidence of polygyny, and slightly attenuated birth intervals. 6nus shift supports Saucier's (1972) conclusion drawn from cross-cultural data that short postpartum taboos tend to be associated with more intensive agricultural women's roles.

316

WElL ADDENDUM

The following addendum is based upon field research in The Gambia in February 1983 which included four days of intensive interviews at Bantunding, Wuli District. Increased Agricultural Intensification

The Mandinka of Wuli Since July 1976, the people of Bantunding have nearly abandoned the extensive production of Digitaria excelis, mentioned in the text as primarily a women's activity, to return to the production of rice on lower-lying freshwater swamp lands during each wet season. While both genders produce rice on their own separate plots, women's plots constitute the large majority of the hectarage planted. The high labor time and low returns per hectare are virtually the same as those described for upland rice; if Bantunding people are to produce wet rice, they do not have the luxury of using the much more productive tidal-swamps such as those found in the Kiangs. The women, as well as the men, have maintained their production of peanuts along with this net increase in labor during the same season. The reinitiation of rice production appears to be the result of at least three factors: (1) the occurrance of heavier rainfall earlier in the wet season than during the drought period preceding 1976; (2) a decrease in the amount of free grains distributed by the government; and (3) the decrease in the availability of imported rice, a decrease which has, in part, been caused by the deterioration of foreign exchange and the government's credit position in the world market. The latter factor has been accompanied by yearly increases in the cash price of imported rice; by 1983, imported rice cost 155 percent more than it did in 1976. These factors constitute strong incentives for the reinitiation of rice production in Wuli and other areas of eastern Gambia. Women have borne the majority of the burden of the increased intensification of agriculture in Wuli.

The Mandinka and Other Ethnic Groups in Other Areas of The Gambia Two other trends are involved in increasing agricultural intensification in other areas of The Gambia among the Mandinka and other ethnic groups. In the Basse area of the Upper River Division, which constitutes eastern

AGRICULTURAL INTENSIFICATION

317

Gambia, and the McCarthy Island Division just west of the URD, many Mandinka and Fulani communities and several Wolof and Soninke (i.e., in Gambia termed "Serahuli") communities have committed increasing land and labor resources to the production of rice through participation in dryseason rice production in addition to wet-season rice production. The first source of this intensification is a national government irrigation program which involves pumping fresh water from the river. Under the program, one or more irrigation societies are formed locally for the purpose of serving as a structure through which extension services are delivered and credit for diesel fuel for the pumps and tractor plowing is organized. Ideally, the irrigation program results in two irrigated crops per year on the same land and in the sale of half the crop each season to pay the loans and to increase cash income for the producers. However, because of such problems as late delivery of fuel and other inputs by the program, only 10 percent of the land under irrigation produces two crops per year. The single irrigated crop produced on the land is usually only produced in the dry season. Even then, yields are often lower than intended because inputs are delivered late as well as because of labor-scheduling conflicts in some areas. The lower yields are thus often too low to provide a surplus for sale, and thus the increased cash income for participants is small or non-existent, the greater availability of rice in the rural market is minimized, and the participants often do not have the cash to pay the program loans. The failure to pay loans for one or more seasons has led in many cases to the withholding of services by the program and thus to non-participation in the irrigation program for one or more seasons-a factor in the small number of cases of double-cropping irrigation. Notwithstanding these problems, a large number of ·communities are involved in single-season irrigated rice production, and while men do participate in such production on either their own plots or on women's plots, the large majority of the labor involved is that of women. The expansion of irrigated rice production has been more than matched by the expansion of rice production during the dry season using "traditional" rice production methods to produce two crops per year. The technical basis of this expansion was provided by irrigation rice production programs sponsored by the Republic of China and the Chinese People's Republic that began in the middle 1960's. The programs introduced several ninety-day rice varieties that were well-adapted to the day-length characteristics of the dry-season; "traditional" wet-season ninety-day varieties were not appropriate for use in the dry-season. In some areas of The Gambia, low-lying lands close to the river and near to its feeder streams (i.e., bolong's) are covered by fresh water throughout all or most of the dry season. Most of the hectarage involved is in the McCarthy Island Division, especially on the north bank. A high proportion of these lands have long

318

WElL

been used to produce wet rice in the wet-season. With the introduction of the new Chinese rice varieties, a high proportion of those communities with access to such lands began producing rice on the same lands in the dryseason. In some cases, the communities also produce wet rice in the wetseason on other lands as well as on these double-season plots; in some other cases, the plots used in the dry-season are only used in that season, and other plots are used for wet-season production. The expansion of the production of rice in two seasons was greatest during the decade of the 1970's. The factors favoring the addition of dry-season rice production include those mentioned in the above discussion of Bantunding's reinitiation of wet-season production. The 1983 research was too brief to be definitive, but it appears that wherever communities are involved in the production of rice in both seasons, women do not participate in peanut production. Labor hours and returns to labor in each season appear to be superior to those described for upland rice production but inferior to those for tidal-swamp rice production. As is the case with dry-season irrigation rice production, both men and women participate in the production of rice during the dryseason with women providing labor for all operations and men for the making of seedbeds and/or assistance in harvesting. As a general rule, women thus provide the majority of the labor in the dry-season production period and nearly all labor in the wet-season period. The addition of the new dry-season production has primarily intensified women's roles in agricultural production.

Household Form: Nucleation During the period since 1976 at Bantunding, the trend towards the formation of nuclear family units has continued. In nearly all cases of marriages to which children have been attached through birth or adoption, the marriages have become the basis of nuclear family production units. Most of these new nuclear family units are still resident in the household area in which the husband was born. However, a large majority of the nuclear family production units that included young parents in 1976 have left the husband's natal residential area and formed their own compounds. The age of the children of a nuclear family production unit, i.e., being old enough to undertake some production and food-processing responsibilities, appears to be a factor in how long such units remain in the same compound with the husband's close relatives. In both cases, the trend towards nuclear family production unit formation continues at different stages of physical independence.

AGRICULTURAL INTENSIFICATION

319

Migration Colvin, eta/. (1981) predict that migration in Senegambia will be increasingly characterized by longer-distant, longer- term international migration, and the data from Bantunding supports that prediction. In 1976, only three people, all males, from Bantunding had migrated outside Senegambia and western Mali, and all had been gone for more than one year. One of the men had gone to Paris and two had gone to Sierra Leone; only one was unmarried and childless. By 1983, nineteen men were away having migrated long distances to other African countries and Europe. Ten of the men were in Spain or Germany, and nine of the men were in the Ivory Coast, Nigeria, or Sierra Leone. Six of them are married and have at least one child; the large majority of participants are thus younger men with fewer family responsibilities. Most had sent some money home, but the transfer payments were small in comparison to the money, food, or labor they might have provided to their natal families or wives and children if they had been home. The largest payment for a year mentioned was D 10 or $4.00, and even if this amount is tripled to adjust for a felt need on the part of informants for secrecy about income, the amounts involved in transfer payments are miniscule. When long-term, long-distant migrants return, they tend to use most of their savings, if any, for such investment purposes as setting up small retail shops and tailoring shops. Those involved in this type of migration were already participating in shorter-term migration within Senegambia in 1976 or would probably have been if they had been old enough at that time. Thus the new migration pattern appears not to represent a net increase in those migrating so much as it constitutes a process having a more dramatic impact on their production units than the older migration pattern in terms of the loss of labor throughout the production year for two or more years. Informants indicated that the wives of long-term migrants have the least access to goods and labor assistance in comparison to the home production units of unmarried migrants which share labor and goods within the units. This is the case for the wives even when they live with the natal production unit of the husband; most married migrants had already formed production units independent of their natal production units, and their wives were thus expected to support themselves and their children. While the new trend in migration affects both natal production units and those of wives left behind, the impact is most negative for the latter. Moreover, if the proportion of young men involved increases, fertility may decrease somewhat through longer abstinence periods. Conclusion The value of these data and the conclusions drawn from them is limited by the short period of the supplemental research and by the lack of

320

WElL

systematic measurement and quantification in most instances. However, they suggest that the trends favoring the maintenance of high levels of fertility and supportive of higher levels are continuing to develop among the Mandinka of The Gambia as well as amongst other ethnic groups. The trends are generally most negative for women in the sense that the burdens of increasing agricultural intensification are primarily being carried by them and in the sense that the burden for the increasing recruitment of labor through birth in the increasingly predominant nuclear family production units is seen as their responsibility. Even in the cases of the replacement of some human labor with .machinery, i.e., tractor plowing in dry-season irrigation production and some non-irrigation rice production, the replacement is primarily in the context of new and additional labor responsibilities allocated to the women as well as man. The majority of the situations of increased agricultural intensification, as in Bantunding, involve no substitution of machinery for the new labor for either gender, and thus these changes, whether labeled as "development" or not, are maintaining pressures on the predominant independent production units to respond through intentional actions affecting the proximate variables discussed in the text which support high and increasing fertility.

16

Temne Fertility: Rural Continuity, Urban Change, Rural-Urban Differences, and Public Policy Problems Vernon R. Dorjahn

Introduction This paper analyzes rural fertility change, urban fertility change, and rural-urban fertility differences among the Temne of Sierra Leone over the period 1963 to 1976, and interprets the results in light of problems faced by formulators of public policy for such high fertility populations. Nearly twenty ethnic/linguistic groups are found in Sierra Leone. The largest are the Temne and the Mende, each comprising roughly 30 percent of the population. The indigenous population of what is today Sierra Leone was joined, beginning in 1789, by repatriated ex-slaves from Great Britain and her Colonies in the New World. Freetown became a naval base and site of a court to try slave traders after the abolition of the trade made enforcement necessary. Many of the slaves freed by the court in Freetown remained there and these ''liberated Africans'' plus the earlier settlers provided the basis for the population subsequently labeled Creole. As the first to be educated in Western type schools and skills, the Sierra Leone Creoles migrated widely up and down the West African coast and engaged primarily in trade and mission work. Closer to home the Creoles, and British interests, expanded into the area near Freetown. In 1896 a British Protectorate was declared and boundaries with French Guinea and Liberia were later settled. By 1900 a functioning British colonial administration had been established. The 321

322

DORJAHN

introduction of the ministerial system in 1953 accelerated the independence process, which terminated on April27, 1961, Sierra Leone's Independence Day. Sierra Leone has been a one-party state since June 1978. Nineteenth century census attempts were limited to Freetown and the surrounding area, some 250 square miles which had constituted a "crown colony.'' The population of the protectorate was estimated at various times between 1901 and 1948 and these estimates, added to the colony figures, showed a population increase from 1,024,278 in 1901 to 1,858,275 in 1948, apparently a rather low rate of growth. In April, 1963, the first national enumeration of population was undertaken and yielded a population of 2,180,355. Based on a post-enumeration survey, an undercount of c. 3.5 percent was indicated. Results of a late 1974 census have not been and apparently will not be published. Dow and Benjamin (1975:429) have estimated total populations of 2,550,376 (1970), 3,130,816 (1980), and 3,987,000 (1990); these estimates are more likely to be low than high. By 1980, if not some years before, the annual rate of growth had passed 2 percent. Tsui et al. (1979:Table 1) project 3,289,000 (1980), 4,457,000 (1990), and 5,768,000 (2000); hopefully, these will prove to be high. The urban segment of the population has increased in the past and will continue to increase in the future at a faster rate than the rural segment owing to rural-urban migration. Therefore, any rural-urban differences toward lower fertility and/ or mortality will have increasing significance as time goes on. Agricultural produce, principally palm kernals, was Sierra Leone's basic export through the colonial period until the economic collapse of 1930. Fortunately, a minerals industry-gold, platinum, and iron initially-was begun at that time and by 1955 the export of minerals exceeded agricultural exports, with diamonds and iron the most important commodities. Today, perhaps 75-80 percent of the country's labor force is involved in agriculture and related occupations; roughly two-thirds of their production is subsistence. Rice, both upland and swamp, is the major food crop and is supplemented by cassava, peanuts, sorghum, corn, millet, bananas, and so on. Trade balances in the 1970s were usually unfavorable and recurring budgetary crises have slowed development efforts. The magnitude of Sierra Leone's development task in light of the approximated demographic situation is suggested by the projections calculated by Tsui et a/. (1979) with respect to education, labor force, health, food, and family planning. World Bank figures show average annual growth rates for Sierra Leone, 1970-79, of 2.6 percent for population and -1.2 percent for per capita (real) GNP. More will have to be done but there will be less with which to do it. The Physical Quality of Life Index (PQLI), a simple measure of the standard of living, is 27 for Sierra Leone (mid-1981) compared with

TEMNE FERTILITY

323

32 as the average for the African continent, 65 as a global average, and 94 as the North American average. The outline of this presentation is as follows. In section 2, I discuss data collection by means of survey enumerations in the Temne context. Background information on Kolifa Mayoso and the town of Magburaka is provided in relation to a discussion of the use of age-calendars to date vital events. In section 3, I present the data on Temne fertility in terms of age proportions, child/woman ratios, maternity ratios, total maternity ratios and the mean number of live births and living children per wife per decade of marriage and show that rural fertility has changed imperceptibly over this period while urban fertility, which approximated rural fertility in 1963, has fallen. Employing Bongaarts' (1978 and 1982) schema in section 4, I discuss the relative importance of the various intermediate variables in accounting for changes and differences in Temne fertility. Declining urban relative to rural fertility appears to be explained primarily by a decreasing proportion of married women in Magburaka. In section 5, I discuss cultural factors accounting for the persistence of high fertility in this population. Implications for planning and policy-making are addressed in the final section.

Methodology Data were collected by the author in 1963 and 1976, 1 with the help of Dr. Antony H. Smith in the urban area in the latter year. The rural data are derived from lOOOJo enumerations of what was formerly the chiefdom of Kolifa Mayoso. The urban data are derived from a one in five block sample of the provincial town of Magburaka, some fifteen miles north of Kolifa Mayoso. Total population figures for rural Mayoso were 2,708 in 1963 and 3,017 in 1976; for urban Magburaka the totals for the blocks enumerated were 1,278 in 1963 and 1,821 in 1976. The general lack of complete registration systems for vital events in African countries has compelled analysts to utilize census/survey data. 2 It seems unlikely that adequate registration systems will be introduced and maintained in the near future and thus dependence on census/survey data will continue. Census surveys yield retrospective data and the potential usefulness and validity of such data are essentially dependent upon the accuracy and completeness of informants' /respondents' memories. The major concern in data collection is thus to ensure and to assist informant recall. In the quest for accuracy and completeness of data, informants' recollections must be aided particularly with respect to dating vital events. It is necessary for the investigator to acquire a knowledge of local happenings and events in many spheres if accuracy in aging and dating is to be achieved.

324

DORJAHN

My own experience in rural Kolifa Mayoso will illustrate ·some of the difficulties encountered in collection of the data upon which this study is based and some of the steps taken to minimize and overcome these difficulties. Kolifa Mayoso, one of the smaller Temne chiefdoms, is located along the west bank of the Pampana River in central Sierra Leone. Through most of the colonial period it was administered as a separate political entity with its own "Paramount Chief." Following the introduction of the Native Administration System, colonial authorities strove to amalgamate chiefdoms so as to have few~ but larger units so that administration might be made more efficient and, later on, so that development would be enhanced. Development was to be fmanced from local revenues, and larger entities would have bigger sums of money available, as well as government funds, the use of and accounting for which could be more carefully and efficiently monitored if there were fewer local government units. In November, 1953, Kolifa Mayoso was amalgamated with Kolifa Mamunta-Rowala, itself the result of a previous amalgamation, whose Paramount Chief and headquarters were located in Magburaka, some fifteen miles away. In 1954-55 there were only four people in the former chiefdom of Kolifa Mayoso literate in English, and almost nobody over the age of ten knew their own age in years. The traditional lunar calendar month names were generally used and most of the cultural beliefs and practices described in the earlier literature on the Temne proved to be still extant (Thomas 1916). Over some months I was able to put together an age-calendar covering roughly 75 years of between thirty to forty absolutely dated events. Most of these events were remembered by many people throughout the chiefdom. Especially valuable were political events such as the deaths and installations of Paramount Chiefs in Kolifa Mayoso and adjacent chiefdoms, and happenings, including the opening of a mission ·church or dispensary, the arrival of the railroad tracks at three different villages, the gold boom on the Pampana River, the diamond rush, the influenza epidemic, the locust invasion of 1928, three large Poro society initiation sessions, and so on. Interviewing subsequently related births, deaths and so forth to these events. In 1962-63 and again in 1976 I was able to add later events such as the blacktopping of a section of the main road, the opening of a feeder road, the construction of a small foot bridge, the Independence Day celebrations, subsequent elections, my own 1955 census visit and so on. In several instances I was surprised to find event X remembered by only a few individuals when I was certain it would be generally recalled, while event Y, which I saw as ephemeral and unlikely to be marked, was widely recalled. Experience over the first weeks of the 1955 census taught me that the basic age-calendar could and should be supplemented by additional datable

TEMNE FERTILITY

325

events well remembered in one or a few neighboring villages. The coming or going of a particular school teacher, missionary, extension agent of the Aureol Tobacco Company and so on, a particular court case (most often a land dispute), the killing of a leopard outside a village, and others all proved useful on this more limited basis. A third category of events proved useful with members of one patrilineage, or a group of adjacent households. The key approach was to cross-check dates for births, deaths, marriages, divorces and so on, both within and between neighboring households, a progressively boring but necessary procedure. The Temne, like most agrarian peoples in West Africa, do not value privacy as we do. A Temne is rarely alone, in a spatial sense. Houses are crowded by American standards, and most work is done in company with others. Conversation is ubiquitous and each person knows a great deal about those living with or near him or her. Recalling past events, who was present, why they were there and so on, is a major activity in which those of all ages participate. Initially I was surprised at the lack of reticence to discuss "personal matters" (our view) such as the presumed causes of infant deaths, who initiated a divorce action, or how a birth could occur three weeks following a marriage. Quickly I found that census interviews would be conducted in public with a participant audience of family and kin group members, neighbors and passers-by whether or not I wanted it that way. To my pleasure and satisfaction I heard listeners prod an informant, question a statement, challenge a date and suggest that another person be asked who was right. Dating a past event in relation to other past events-i.e., using an age-calendar-was precisely the way Temne approached the problem; I added only absolute dates retrieved from government records, diaries, the published literature and so on. The above demonstrates that the range of background information useful and necessary in assigning dates and ages is formidable and normally available only to a few local residents and a student of life in a particular community or area who expends considerable time, effort and money acquiring such knowledge. It is precisely to this kind of limited yet intensive research that ethnographic field methods are best adapted and it is in such studies that anthropology in general and individual anthropologists can make their best contributions to population studies. The town of Magburaka is a "new town," one which came into being during the colonial period, around 1900. An estimate for 1927 gives a population of only 348 (Dumbuya 1973: Table 2). A 1946 estimate, based on tax listings multiplied by a factor presumed to approximate the mean number of inhabitants per house, shows 4,500 inhabitants (Atlas of Sierra Leone 1953:10). The author's general census of November, 1962, yielded a population of 5,680, while the enumeration of the Sierra Leone census of

326

DORJAHN

1963 gave a' figure of 6,371. My tentative impression in late 1976 was of a population ranging upwards of 7,500. Elsewhere (Dorjahn m.s.) I have described the growth of the city and the important role migration played in this growth. Over the years Magburaka attracted migrants from all the forty-odd Temne chiefdoms and all districts of Sierra Leone. Many moved on, of course, but over time a core of permanent townspeople has developed. The heterogeneous origin of Magburaka's population and the comparatively high turnover of urban residents made the construction and employment of an age-calendar and the extensional devices and strategies described above less rewarding. In Magburaka I opted for the lOOOJo enumeration of one block in each five, rather than every nth house on the assumption, only partially correct as it turned out, that neighbors would have knowledge of neighbors and help out in the type of group interview situation I knew from research with rural Temne. A number of points should be made in this regard. First, the heterogeneous origins of rural-urban migrants rendered an age-calendar based on events even in the several adjacent chiefdoms of only limited use. Many migrants had lived in the town for only brief periods and had only a partial and temporally shallow knowledge of past events there. Second, and considering only the Temne majority, there was less knowledge of neighbors and co-residents, and, where there was knowledge, there was greater reticence to give it voice. Interview sessions attracted neighbors and passers-by, but in the urban context they were frequently asked or even ordered to go. Only in Magburaka did I and my assistant hear the whispered "I'll tell you later when we're alone," indicating a view of private and privileged information different from the rural Temne norm. Third, and on the plus side as an important compensating factor, more urban residents knew their own ages, the dates of births, deaths, marriages and so on. In sum, I believe the dating and aging to be more accurate for the urban than the rural series, while, on the other hand, omissions/underenumerations are more frequent in the urban context. The effect would be to depress, slightly, the reproductive indicators for Magburaka.

Temne Fertility The breakdown of age/sex data, usually in five or ten-year groups, provides a useful first look at reproductive performance. Data in Table 1 show that in rural Mayoso the modest reduction in the 0-4 category is more than offset by the increase in 5-9's. The rural population is somewhat younger in 1976 than in 1963. However, the population of urban Magburaka shows a sizable decrease for 0-4's and a modest decrease for 5-9's, giving an overall

TEMNE FERTILITY

327

decrease of roughly 50 per thousand for the age group 0-9. Rural-urban differences, 1963 and 1976, show reversals for all three age groups from higher urban figures to lower urban figures. TABLE 1 TEMNE AGE PROPORTIONS (PER 1000)

Population Rural Mayoso 1963 1976 Urban Magburaka 1963 1976 Rural Marampa* 1960 Urban Lunsar* 1960 All Temne** 1963

0-4

CbUdren Aged: 5-9

0-9

191 182

155 178

346 360

206

164

159 150

365 314

240

146

386

133

146

279

179

146

325

*Mills (1967) **Davis (1973: Table 4)

Age-proportion data must be used with caution where small populations are concerned, owing to the effects of differential migration and of variations in mortality as well as fertility and age misstatement. In this instance the major explanatory factor is probably the age composition of the inmigrant populations. In Magburaka there has been, in the interval between 1963 and 1976, an increase in the number of secondary school students and would-be students which has bulked up the 10-19 age group. In Mayoso the establishment of three government primary schools has brought in a number of students aged 5-9 (actually S-14) from adjacent chiefdoms with fewer schools, students who live with relatives and friends in Mayoso. In Tables 1, 2 and 3 I include data on the town of Lunsar, an iron mining center in 1960, and on some peri-urban villages in rural Marampa chiefdom. Relatively, there are decidedly fewer 0-4's in Lunsar and more in the peri-urban but rural villages in comparison with Magburaka and rural Mayoso. Living accommodations in a mining town may have required the boarding of children in nearby villages. The "All Temne" figures are based on data from the 1963 government census. The percentage of any

328

DORJAHN

population aged 0-4 can be divided by five to give an approximation of the crude birth rate. This gives an underestimation, of course, since those who die in the first five years of life are not reckoned. The higher are infant and child mortality, the greater the under-estimation. Thus, the crude birth rates so derived, for rural Mayoso, 38 (1963) and 36 (1976), and for urban Magburaka, 41 (1963) and 33 (1976), should be viewed in light of the child mortality (0-4) rates of 393 (1963) and 364 (1976) for Mayoso and 294 (1963) and 245 (1976) for Magburaka. TABLEl TEMNE CHILD/WOMAN RATIOS (PER 1000)

Population Women Aged 15-44 Rural Mayoso 1963 1976 Urban Magburaka 1963 1976 Rural Marampa* 1960 Urban Lunsar* 1960 Women Aged 2.0-49 RurBI Mayoso 1963 1976 Urban Magburaka 1963 1976

0-4

ChUdren Aged: 5-9

0-9

909 873

739 854

1,648 1,727

886 808

684 738

1,570 1,546

1,034

626

1,660

483

528

1,011

893 924

726 904

1,619 1,828

1,019 1,099

787 1,004

1,806 2,103

*Mills (1967)

Child/woman ratios are a somewhat better approximation of fertility performance since instead of relating the number of children 0-4, 5-9 or 0-9 to the total population, they relate them only to living women in the childbearing years, usually 15-44 or 20-49 (Kuznets 1974). These data are fouild in Table 2. Looking first at the figures derived using women 15-44 in the denominator, it is apparent that for both rural Mayoso and urban Magburaka there are reductions for 0-4's, and increases for 5-9's. The

TEMNE FERTILITY

329

relative magnitude of these differences is such that for 0-9's there is an overall increase in Mayoso and decrease in Magburaka. It appears that fertility is lower in the urban than the rural area in both 1963 and 1976 and that the difference is widening. The Lunsar data provide decidedly lower ratios for the three age groups of children, not a surprising situation for a mining town. The rural Marampa villages have decidedly more 0-4's and fewer 5-9's. However, looking at the figures (Table 2) derived using women 20-49 in the denominator, it is apparent that for both rural Mayoso and urban Magburaka there are increases for 0-4's and even larger increases for 5-9's and 0-9's. It appears that fertility is higher in the urban than in the rural area both in 1963 and 1976 and that the difference is widening. Clearly it makes a difference in this case which age group is used in the denominator. Child/woman ratios, whatever age groups are used, are as responsive to variation in age composition owing to age/sex selective migration as they are to fertility differences and age misstatement. · The 0-4 child/woman ratio, divided by 5, gives an approximation of the general fertility rate, defined as the number of births in one year per 1,000 women of childbearing age, usually 15-44 or 15-49. Such approximations are underestimates since mortality in the first five years of life is greater than the mortality of women in the childbearing years, and this must be kept in mind in comparative work. 3 This procedure yields estimated general fertility rates for Mayoso of 178 (1963) and 174 (1976) and for Magburaka of 177 (1963) and 161 (1976), all using women 15-44 in the denominator. Given higher child mortality in Mayoso (see below), the rural figures are surely somewhat higher. The maternity ratio, the mean number of live births ·per woman ever married (regardless of age at the time of inquiry) is a reproductive measure as responsive to differences in the age distribution of mothers and the mean number of married years as it is to fertility differences. The maternity ratios in Table 3 indicate a possible, small increase in rural fertility, 1963 to 1976, a definite increase in urban fertility and a reduced urban fertility deficit. The deficit is probably real, however, since the married female population is younger in the urban area. Possibly differences in the age distributions of mothers account for some of the difference evident between rural Mayoso and rural Marampa. The 1913 data collected by Thomas, yielding a maternity ratio of 2.6, suggest remarkable continuity; my own 1955 data from rural Mayoso also yielded a maternity ratio of 2.6. The total maternity ratio, the mean number of live births per living woman aged 45 and over, is a measure of completed fertility. The data in Table 3 indicate that fertility so measured is decidedly higher in rural Mayoso both in 1963 and 1976, but that there is no evidence of reduced fertility in either location over that time period. This suggests that if there is, in

330

DORJAHN TABLEJ

TEMNE MATERNITY AND TOTAL MATERNITY RATIOS (PER WOMAN) Matemity Ratio

Total Matemity Ratio

1963 1976

2.6 2.8

3.6 3.6

1963 1976

2.4

1.9

2.4 2.5

1960

4.1

5.6

1960

2.5

4.7

1913

2.6

n.a.

PoJ!ulation Runl Mayoso Urban Magburaka Runl Manmpa* Urban Lunsar* Runl Temne**

•Mills (1967)

••Thomas (1913)

fact, a fertility reduction among younger women that it is too recent to be reflected by a measure of completed fertility. Total maternity ratios were decidedly higher in rural Marampa than rural Mayoso and in urban Lunsar than urban Magburaka. My 19SS census in rural Mayoso showed a total maternity ratio of 4.0. In Table 4 data are grouped by age of mother (at time of inquiry) and standardized per ten years of married life for all wives ever married to the men enumerated. For rural Mayoso there are no changes from 1963 to 1976 for the total figures with respect to either living children or live births (mean number per wife per decade of marriage). There may be change with respect to age groups 10-19 and 20-29, although the break between them is perhaps open to greater variation than any other owing to age misstatement/assessment. For urban Magburaka there are apparent fertility decreases from 1963 to 1976 for total figures and for both measures. There are no striking differences between age groups of mothers/wives with the possible exception of 40-49. Urban fertility appears to be somewhat lower than rural fertility for 1976 but not for 1963. Effects of Intermediate Variables Recent work by Bongarts (1978, 1982) seeks to demonstrate that fertility differences between populations are largely the result of variation in four

331 TABLE4 LIVE BIRTHS AND LIVING CHILDREN PER DECADE OF MARRIAGE

Population Age Group

Wives

Mean # of Living CbUdren per wife per Decade of Marriase

Mean # of Live Births per wife per Decade of Marriase

1963

Runl Mayoso 50 & over

40-49 30-39 20-29 10-19

158 176 199 207 48

1.0 1.3 1.7 2.0 1.0

1.6 2.0 2.3 2.4 1.6

Total:

788

1.4

2.0

40-49 30-39 20-29 1019

24 44 100 138 31

.7 1.0 1.8 2.2 2.0

1.7 2.5 2.6 2.0

Total:

337

1.5

2.1

Urban Magburaka 50 & over

1.1

1976 Rural Mayoso 50 & over

40-49 30-39 20-29 10-19

191 215 255 166 61

1.0 1.4 1.7 2.1 .9

1.7 2.1 2.3 2.7 1.3

Total:

889

1.3

2.0

40-49 30-39 20-29 10-19

28 44 119 140 39

.6 1.6 1.8 1.8 1.7

2.2 2.5 2.6 2.1

Total:

370

1.3

2.0

Urban Magburaka 50 & over

1.1

332

DORJAHN

intermediate variables: (1) proportions married among females, (2) contraceptive use and effectiveness, (3) prevalence of induced abortion, and (4) duration of postpartum infecundability. Three other intermediate variables (5) fecundability (frequency of intercourse), (6) spontaneous intrauterine mortality and (7) prevalence of permanent sterility, he argues, either have less effect on fertility levels or vary less frequently/significantly between populations. In particular instances, he recognizes, one or more of the second three may be important. Looking at the Temne data in light of Bongaarts' schema, the situation can be summarized as follows 1. Proportions married among females differ little for rural Mayoso between 1963 and 1976, they drop for all age groups for urban Magburaka and are significantly different in 1976 between rural and urban series. This factor thus accounts for some differences/changes in fertility. 2. Contraceptive use and effectiveness probably have an insignificant effect, if any. At least in terms of Western techniques, these are noncontracepting populations. Traditional techniques were/are known but their effectiveness is either unknown or cannot be demonstrated. 3. Induced abortion, also, is extremely rare/nearly nonexistent. It can have little or no effect. 4.

Duration of postpartum infecundability probably does vary but is diffic;:ult to estimate and impossible to measure in the Temne context. My best estimates of weaning age are 24 months in the rural area and 18 months in Magburaka, based on observation and assessment of birth intervals on a selective basis. 4 There was probably little if any change in rural Mayoso, 1963 to 1976; conceivably there has been a slight reduction in the urban area. This factor certainly has an effect but its magnitude is difficult to measure.

Changes in nuptiality, as indicated by the percent of females in a given age group who are married at the time of an inquiry, can be examined with the data in Table 5. There is little if any change in rural Mayoso, 630Jo (1963) to 600Jo (1976), and this is easily accounted for by the in-migration of female primary school students in their late preteens and early teens. Conversely, in urban Magburaka there has been a sizable overall reduction from 670Jo to 420Jo, a reduction to which all age-groups contributed. The explanation of this change is complex but will be briefly outlined.

TEMNE FERTILITY

333

TABLES TEMNE PROPORTIONS MARRIED

Population Age Grou2 Rural Mayoso 50 & over

1963 Women

f

Married %

Women

Married

1976

f

%

40-49 30-39 20-29 10-19

224

147 203 230 141

95 113 161 182 46

42 77 79 79 33

241 176 230 189 221

102 134 194 146 58

42 76 84 77 26

Total:

945

597

63

1,057

634

60

40-49 30-39 20-29 10-19

56 27 91 140 106

14 24

74 47

130 29

25 89 82 93 27

120 245

14 29 72 88 44

19 62 69 73 18

Total:

420

272

67

591

247

42

Urban Magburaka 50 & over

15

lOS

First, there has been, over the period 1963-1976, an expansion of secondary school facilities and a net in-migration of females in their teens and early twenties as students, would-be students and companions associated with the comparable influx of males. Second, possibilities for single women to subsist in the urban area have increased. Trade, preparing and selling cooked food and so on are realistic possibilities even for young women. Older women, usually widows owning houses, take in boarders, including groups of primary and secondary school students, and continue to live in town in female-headed households. 5 While there is no way to quantify the effect, it is fairly clear that "consensual unions," "live-in" arrangements and so on have become more frequent and these, too, enable more technically single women to remain in town. Third, there has been a loosening of kinship bonds, especially between those not co-resident in the same household or village, which makes female urban-rural migrants less welcome and less likely to receive help if they return to natal villages. The value of females in rural areas still rests fundamentally on their contributions to farm and household labor. Rural informants raised several points in this regard. Older urban women who had not done farm work for many years, some argued, were not in adequate physical condition for physically

334

DORJAHN

heavy farm work; their hands blistered, their muscles cramped and so on. Younger women, often urban-raised if not urban-born, were untrained and hence less useful for farm work. 6 Furthermore, urban women who had been to school tended to look down on and ridicule farm work, denigrate living conditions and diets in rural households and make fun of and lord it over their rural counterparts. Too often the urban-rural female migrant "makes trouble" and rather than helping with household chores must be waited on and looked after. Rural informants would have one believe that such women time their movements so as to avoid the periods of peak farm labor input: "She stayed in tow~ until after the rice was harvested, let her go back there now." Also, in rural areas extended households are becoming rarer and mean household size is decreasing;7 the smaller the household the less food surplus is produced and the more difficult it is to provide for visitors, short or long term, especially for those who can't or won't pitch in to help. Finally, there has been a modest aging of the Magburaka population, including the presence of more widows who inherit houses, carry on businesses and continue urban residence. For them the rural retirement option is less desirable and less often necessary than for their counterparts a decade or two earlier.

TABLE6 AGE DISTRIBUTION OF WIVES Population Ase Grou2

f

Rural Mayoso SO It over 40-49 30-39 20-29 10-19

95 113 161 182 46

Total: Urban Magburaka 50 It over 40-49 30-39 20-29 10-19 Total:

1963

%

16 19 27 30 8

272

102 134 194 146 58

%

16 21 31 23 9

634

597 14 24 75 130 29

1976

f

5 9 28 48 11

14 29

72

88

44

247

6 12 29 36 18

335

TEMNE FERTILITY

Data on the age distribution of married females in Table 6 show little in the way of changes from 1963 to 1976. There may be modest aging trends in both locations. Rural-urban differences, however, are consistent for the two years and are striking. The differences in the percent of all wives 40 and up show a marked rural preponderance, 350Jo to 14% in 1963 and 37% to 18% in 1976. As regards the married female population 29 and under, there is a striking rural deficit, 380Jo to 59% in 1963 and 32% to 54% in 1976. Clearly the married female population is older in the rural area with consequent effects on various fertility measures. At least one of Bongaarts' three secondary intermediate variables, sterility, is also a factor in the Temne context. Indeed, informants conveyed feelings of concern for what demographers call primary and secondary sterility of females. The data in Table 7 indicate a possible numerical basis for informants' concerns. Considering only women who had been married for 10 or more years at the time of inquiry, the portion having had 0 live births is consistent at 11-12% for both years (1963 and 1976) in both rural and urban populations. Both thus fall within Nag's (1968:121 and Table 62) "high range of sterility."

TABLE7 EVER MARRIED WIVES HAVING 0 OR 1 LIVE BIRTH AFTER 10 OR MORE YEARS OF MARRIAGE 1963

1976

Population Age Grou2

f

Rural Mayoso 50 & over 40-49 30-39 20-29 10-19

21 13 16 3 0

17 10 12 8

12 12 17 8 0

10 9 13 22

17 20 12 3 0

13 8 9

11

15 20 24 2 0

10 13 16 6

53

12

49

11

52

11

61

12

5

26

19 9

3 4 4 0

16 19 8

7 6 9 0

26 16 10

7 2 6 0

26

12

14

12

24

12

16

8

Totals: Urban Magburaka 50 & over 40-49 30-39 20-29 Totals:

OLB

4

5 0

14

%

f

lLB

%

f

OLD

%

f

1 LB %

5

7

336

DORJAHN

Bogue (1969: Table 18-31) presents data on the "Proportion of Women Aged 40 or Over Who Have Borne No Children"; the African figures cited range from 5.40Jo (Kenya) to 32.1 OJo (Gabon). Looking only at women aged 40 and over (and married at least 10 years) the Magburaka figures are 230Jo (1963) and 200Jo (1976); the figures for rural Mayoso are 120Jo and 120Jo respectively. In developed countries Bogue (1969:726) states that when childlessness rises above 100Jo in couples, one infers either voluntary childlessness (resulting from the use of contraceptives) or very irregular exposure to childbearing as explanations. The Temne populations are noncontracepting and primary sterility is usually the result of involuntary physiological conditions. The African data have been reviewed by Romaniuk in the volume edited by Brasset a/. (1968: Table 2.17). Secondary or "one birth" sterility may result from anything causing primary sterility (childlessness) or a range of infectious conditions developing postpartum. Conditions of birth in rural areas seem to foster a noteworthy incidence of secondary sterility, 11-120/o. Comparable incidences for urban Magburaka of 120Jo (1963) and 80Jo (1976) suggest a decline probably owing to the increasing proportion of births taking place under improved conditions. The 1963 data show that 120Jo of those women married 10 or more years had contributed no live births and another 11-120/o only one live birth per woman. The remaining 76 to 78 women in every hundred contribute sufficiently so that overall fertility indicators show a moderate, but not high level of performance. TABLES INCIDENCE OF POLYGYNY AMONG MARRIED MEN

Married Po2ulation Men Rural Mayoso

1963 1976

374 414

Urban Magburaka

1963 1976

179 173

Percent of Married Men witb N wives >4 4 3 2 1

Wives per

100

Husbands

60.7 61.8

28.1 26.3

7.5 8.9

2.1 2.6

1.9 .2

156 153

65.9 67.0

24.0 22.0

8.9 4.6

1.1 .6

0 0

145 133

With respect to the remaining pair of intermediate variables, spontaneous intrauterine mortality and fecundability, the Temne situation is less certain.

TEMNE FERTILITY

337

Late fetal mortality (stillbirth) rates, and the more reliable perinatal mortality rates (Dorjahn 1976: Tables 1 and 2), are not markedly higher for the Temne locations than in comparable areas of West Africa and thus these mortality differences cannot have a major influence on fertility differentials. Fecundability (frequency of intercourse) is determined both by desire (not necessarily mutual) and by opportunity. Even within the married Temne series utilized in this study, the two are inconstant. Opportunity is reduced essentially, I would argue, by the spatial separation entailed by migration (for whatever reason) and the reduced frequency of intercourse on a per wife basis in polygynous households (Dorjahn 1958). Once again I doubt that the periods of separation are any greater for the Temne than for most West African populations (Dorjahn 1975). Similarly, the frequency of polygynous unions, summarized in Table 8, is not particularly remarkable for West Africa. It is evident that, whatever the magnitude of the fertilityinhibiting effect of polygynous marriage, that effect is only slightly greater in rural Mayoso than in urban Magburaka.

Cultural Factors Affecting Intermediate Variables Intermediate variables determine Temne fertility parameters, but cultural variables determine the intermediate variable parameters. The ''why questions'' have always been critical in ethnographic and ethnologic research and it is at this point that anthropology can make a significant contribution to fertility studies. Data presented in the previous section suggest that urban fertility is lower than rural fertility for the Temne and that this results from ·a certainly lower proportion married, a probably reduced period of post-partum infecundability, and a possibly reduced fecundability for the urban segment. The question now becomes one of describing at least some of the cultural factors8 bringing about these rural-urban differences in the intermediate variables. Too many analyses of differential fertility, based on aggregate census data, offer little on this question. Stycos (1963:5) has suggested that most demographers come from disciplines that rarely utilize and are suspicious of "soft data" on attitudes and opinions. One can't deny that there are problems relating informant-stated attitudes and opinions to behavior, past, present, and/ or future. Certainly there are few, if any, oneto-one correlations. But this is a weak excuse to abstain entirely from collecting and utilizing such data. It is always possible, and in the context of fertility, migration, and several others, I think it likely, that such "soft data" can tell us more than inferences drawn from otherwise sterile mathematical manipulations of aggregate data. If you want to find out why people did something you start by asking them why they did it.

338

DORJAHN

A recent article by Dzegede (1981) dealing with rural-urban fertility differences in Ghana, Sierra Leone, and Liberia serves as an example. Under socio-cultural variables influencing fertility, the author considers ethnicity, religion, socioeconomic status, education, migration, and age at marriage. Discussion points out that in specific instances fertility varies with religious preferences/practice, with socioeconomic status, with education, and so on. Depending on the data, correlations may be solid or tentative. Yet aggregate data analysis essentially stops here. It cannot tell us, in either a specific case or more generally, how and why two ethnic units or adherents of two religions show different fertility performance levels. Levels of fertility and levels of education seem negatively correlated in many instances, but what is it about more years of education that depresses fertility? Is it that more schooling leads to elevated socioeconomic status, the validation of which requires acquisition of expensive material goods, conspicuous consumption, and so on, and that to secure status the expense of having (more) children is seen as too great? Is it something in the school curriculum, something that is learned, which is the key factor: an inference drawn from a presentation in biology of Darwin, in geography of the man-land ratio, a discussion of Malthus, of The Population Bomb or something else? "Education," of course, is tied to "income" in most instances and the relative effects of the two are hard to untangle. Hawthorne (1970: 103) comments: "Despite the great detail provided in the most comprehensive studies, the fact remains that the correlations between education and fertility variables still leave the exact causal mechanisms somewhat obscure." At this point a disclaimer is necessary. I have argued the need to tease out the cultural factors determining the intermediate variables in a given instance. It must be stressed that, however disappointing to those who view fertility lev~ls as the result of conscious, rational decision-making, human fertility is not entirely determined by cultural factors of which the actors are aware and which they can express in answering questions (see Schumann, this volume). People are not always rational in their fertility behavior. Even when they are, the same information can lead to different decisions; the calculus is often imperfect, or even mistaken. Hawthorne (1970:58) refers to fertility behavior as " ... an area in which irrationality and nonrationality are so conspicuous • . . " In the widely cited study, Family Growth in Metropolitan America, Westoff et al (1961:14) refer to this position, one they do not hold, in these words: "This position maintains that fertility and family-size behavior are the consequences of sexual behavior, which is essentially nonrational behavior.'' It is necessary, I think, to appreciate the possibility that a portion of "fertility decisions" are accidental, emotional, and non-rational in relation to cultural factors relevant to such decisionmaking (see Crosbie, this volume).

TEMNE FERTILITY

339

Of those who have concerned themselves with the cultural factors affecting fertility, Caldwell (1976,1981) has made the most intriguing suggestion. In the context of explaining the demographic transition he emphasizes change in the intergenerational wealth flow. This flow is largely determined by social considerations, by changes in social structure. Fertility reduction occurs when the "family morality" changes, that is, when the nuclear family hives off from the extended family to become the basic socio-economic unit of production and consumption. Concomitantly, the husband-wife relationship comes to predominate, parent-child relationships take precedence over relationships between either and members outside the nuclear family in question but still within the extended family, while a shift in mutual expectations occurs from what children owe their parents to what parents owe their children. Utilizing the familial mode of production as a basic concept, Caldwell stresses that the family household is usually set to benefit the male heads and other adult married men in a material and prestige sense. "The struggle of the decision-makers to maintain their advantage is normally seen as the assertion of natural rights and as proper behavior; nevertheless, family economic relationships are exploitative and there is potential for conflict and change" (Caldwell 1976:554). Relatively high fertility benefits the decision-makers because wealth flows from the younger to the older generation. Caldwell's characterization fits the traditional, rural Temne situation well and I would argue that among some urban Temne the changes he speaks of have begun to take place, and that this is one important reason for decreasing proportions of married women and lowered fertility in Magburaka. In what follows I utilize data from questionnaires supplemental to my 1976 census inquiries to sort out the reasons underlying continued high fertility among these Temne. These were simple availability samples. Data are presented for married men and women in both rural and urban areas. Tables 9, 10, and 12 also include data from teenage, unmarried secondary school students in attendance at Magburaka schools; these students came from both urban and rural backgrounds and represent a second generation. It is useful, first of all, to look at responses to questions of what advantages and disadvantages there are to having a large number of children. The data are summarized in Tables 9 and 10. The most frequent advantage cited by all six categories was "help to support me now"; rural people stressed help in farming while urban people mentioned help in bringing in money, watching the house and minding younger children while older people worked. Males in all three sets more frequently cited support in old age than females. The "one may prosper" response is a probability argument; the more children you have, the more likely it is that one will prosper, make it

340

DORJAHN TABLE9 ADVANTAGES TO LARGE FAMILIES

Res2onse Support me now Support me in old age One may prosper More will survive Prestige of big family Other No Advantage No response N

Married Magburaka Men Women

37

23

82 9

21

6

2 2 2

2

4

3 100

Married Mayoso Men Women

41

43

25

40

30

12 30

9 9

3 4

7 4 1

13

13

4

2

101

100

9 9 89

1 49

19

2

2

100

Magburaka Students Men Women

2 5

big, be especially generous to the parents and so on. Similarly, the more children one has, the "more will survive," a reasonable response where infant and child mortality are both high; the more surviving children one has the greater potential and probability for help. The traditional high "prestige of a big family" was cited less frequently than I had anticipated, especially for the rural series. That there were proportionately more responses of "no advantage" from the younger school persons may indicate a change of potential significance. The difficulty in supporting, including educating, many children was the most frequently tendered disadvantage (Table 10) in having a large number of children. This was the only significant disadvantage cited by Magburaka people and was well stated by one fifty-year-old shopkeeper: "It is hard to feed, clothe and house them; you must slave for them. Educated children who can't find work return home for you to support as they are unused to hard, physical work.'' A Mayoso woman with five living children put it this way: "You will have to buy woman damages, you will not have any money; feeding, clothing and medical care will take all your money." Rural people also frequently cited the difficulty of training/disciplining many children. Predictably the "no disadvantage" response was more frequent among rural people than urban people. The adult respondents in Mayoso and Magburaka were asked (Table 11) whether they thought their parents should have had fewer, the same number, or more children than they in fact had had. A sizable majority of urban people answered "more" while a smaller majority of rural people gave the same response.

341

TABLE 10 DISADVANTAGES TO LARGE FAMILIES

Resf!onse Hard to support Hard to dlscipHne They won't help now or in old age Other No disadvantage No response N

Married Magburaka Men Women

91 4

100

1 1 1 3 100

100

Married Mayoso Men Women

Magburaka Students Men Women

58 20

57 19

37 7

1 1 5

14

2 2 1 10

4 4

101

100

89

49

61 32

TABLE 11 MARRIED PEOPLE'S RESPONSES TO QUESTIONS ABOUT CHILDREN

Question

Response

Do you think your

fewer more same no respns

parents should have had fewer/the same number/more children, and why? Would you Hke to have more children and why? Is it important

to you to have a son? why/why not?

N=

yes no no respns

N=

yes no no respns

N=

Magburaka Men Women

Mayoso Men Women

69 18 2 100

74 14 1 100

101

42 54 3 1 100

71

64

76

64

11

11

26

3 100

36 0 100

94 5

87 13

1 100

0

100

37 52 7 5

24

33 3 100

96

91 9

1 101 4 1 101

0

100

Magburaka respondents of both sexes who answered "fewer" stressed in answer to the why question, that they would have been better cared for or provided for, had their parents had fewer children. Males in Mayoso

342

DORJAHN

stressed the benefits they might have enjoyed: some or more education, better care, more food, more land to farm and so on, while Mayoso females indicated similar things but also referred to the burdens and difficulties their parents had faced. Of the 69 Magburaka men who wished their parents had had more children, 42 indicated additional children would have helped their parents and made life easier for them, 19 remarked on the assistance they might have had from additional siblings, 5 commented on the added prestige gained by members of big families and 3 noted that many of their siblings had died. Of the 74 Magburaka women who stipulated "more" 52 cited help for their parents, 13 indicated they would have had help from more siblings, 6 mentioned high mortality of siblings and 3 the prestige given large families. Three women who had no siblings cited the Temne proverb that "one finger can't pick up a kernal of rice on the ground." Fiftytwo Mayoso men wished their parents had had more children; 28 cited the help those children would have given the parents, 13 the help they themselves would have received from more siblings, 10 the prestige of big families and 1 the likelihood that he would not now be alone. The 54 Mayoso females responded as follows: 31 help for parents, 17 aid for themselves, and 6 the prestige of belonging to a large family. The same persons were asked whether or not they would like to have more children themselves and why. The data in Table 11 indicate a 2:1 or 3:1 preference for more children and the "yes-more" responses are explained by the same set of reasons as detailed above. Looking at the "nofewer" respondents, both Magburaka series and the Mayoso males stress the difficulties of supporting, raising and educating more children. A Magburaka man with foresight commented: "The world wide inflation affects everyone. It will be difficult to feed and clothe children as the prices of this is [sic] too high." A Magburaka woman, speaking from experience, said simply: "They [children] take too much time, money, and sleep." Isolated responses mentioned that children desert you, won't help you, die and so on. The 33 responses ~f Mayoso women differed sharply from those of the other three series: 21 cited the pain of childbearing, 3 said they were tired from (not of) childbearing while 9 stressed the costs of supporting and raising additional children. The question, "Is it important to you to have a son as heir and why?" brought a "yes - important" response from roughly 9 of 10 people in all four series. The modal explanation for three series, Magburaka females excepted, was to control and administer the family and its property and to carry on the family name. The Magburaka females most often cited help and support for the mother and female siblings including avenging them when that should be necessary. As a Mayoso woman noted, 9 "It is hard for a woman to become the head of a family." Help and support was often

TEMNE FERTILITY

343

interpreted quite crudely, as by this Magburaka man: "I can use him as an instrument to do anything for me." While help and support was the second most frequent response for Magburaka men, it was third for both Mayoso series behind the observation that "daughters leave but sons stay." Implied here, of course, is that sons thus provide more help and support. In the words of one Mayoso women: "All my daughters are married so they can't come and stay with us. Now only my son stays with me." It was sometimes phrased that sons care more for their parents than daughters do. "No -unimportant" responses were more frequent from females than males in both locations; most stressed that daughters can better care for a woman/mother and that daughters are equal sources of support.

TABLE 12 RESPONSES TO QUESTIONS ABOUT OLD AGE

Response

Magburaka Men Women

Mayoso Men Women

Magburaka Students Men Women

Wben you are old and retired do you expect to be supported by one of your children and wby? yes 94 100 IOO IOO 69 4I no 6 0 0 IS 0 7 no response 0 0 I 0 7 2 Wben you are old and retired do you expect to live witb one of your cbUdren wby? yes 70 8I 93 92 32 no 30 I9 7 5 54 no response 0 0 I 3 0 N= IOO IOO IOI IOO 86 Of tbose wbo responded "yes" to tbe previous question, to Uve? son 62 31 63 daughter I 46 0 uncertain 7 4 30

and 17 3I 0 48

witb wbicb do you expect 47 6 39

Responses to the question "Do you expect to be supported in old age by your children and why?" are tallied in Table 12 by sex for each of the series. ''Yes -supported'' responses are heavily dominant with any real deviation being generational. The reason cited in answer to the why part of the question by both married Magburaka males and females was "they owe me since

344

DORJAHN

I raised/supported them" or "they must repay me for raising/supporting them." An undertone of "quid pro quo" is evident. Mayoso people gave more varied replies but for both males and females the modal response was a simple "That is why I had children." As one Mayoso man explained it: "It is in form [sic] of a debt, because I have suffered for them, so they in turn should support me in old age." Others added that "it is custom," or "they owe me," "they must repay me," "I supported my parents and my children should support me" and so on. A few stressed the anticipated inability to support themselves by farming in old age or stated simply, "I will need help.'' Interestingly the secondary school students cited the same range of reasons. The minority of "No - no support" responses can be readily summarized: about half indicated they expected a pension and/or saving enough money from salaries to support themselves. The other half cited the selfishness/ingratitude/limited earning capacity of the young or the fact that their children will need all the money they can make for themselves. While a substantial portion of respondents thus indicated they expected to be supported in old age/retirement by their children, a smaller portion in both Magburaka and Mayoso wanted to live with one of their children in old age. These data, also in Table 12, show that a majority of the school series did not want to reside with one of their children at retirement. There appears to be a slight rural-urban difference. Essentially, those who indicated a "yes" -live with children" response gave the same explanations as to the previous question regarding support. The negative responses fall into three categories: (1) I would embarrass, inconvenience or impose upon them, (2) I would be embarrassed, annoyed or unhappy living with them, and (3) I want to avoid trouble, quarrels or palaver. A Magburaka man phrased it: "I don't want to experience many indignities or total molestation." A Magburaka woman said: "My daughters would bring in lovers and my daughters-in-law are hard to live with in peace." A Mayoso woman predicted: "I would surely suffer from frustration, embarrassment and indignities." A 17 -year-old school boy, perhaps with an eye toward crowded or overcrowded Magburaka housing, wrote: "I would surely see what I would not like to see going on between my son and his wife." A 15-year-old school girl commented: "It is shameful for me to live in my daughter's house. Maybe I'll disturb them a lot. I'd rather live my myself. If she happens to have anything she can send it to me." Those adults who answered "yes" to the previous question were also asked with whom they wanted to live in retirement. Clearly the modal response (Table 12) is with a son, usually the eldest, upon whom the obligation ideally fell in the traditional Temne view of life. Most respondents stressed this obligation, his greater wealth and sense of responsibility and so

TEMNE FERTILITY

345

on. Excluding one Magburaka man who had no son and thus planned to live with a daughter, only women expressed the intent to live with a daughter and most of those were urban women; they stressed that a daughter could bathe them and better care for them in illness. The uncertain respondents said they would live with whomever could better provide for them or loved them the most or was the wealthiest. The fundamental question with respect to the data presented in this section is, quite simply: do these attitudes, presumably indicative in some way of underlying values, influence fertility levels? Implicit, of course, is the question of to what extent fertility levels reflect conscious choices by Temne, and, who is involved in any such decision making? Neither question has a simple, readily demonstrated answer. I interpret the data in Tables 9 and 10 and the reasons given for responses as indicating a pressure for higher fertility; the advantages of a large number of children, whatever is meant by a large number in each individual case, outweigh the disadvantages. It must be stressed that even though people see some difficulties and disadvantages to a course of action, those won't prevent them from trying it if they feel the advantages are worthwhile. A majority of respondents (Table 11) wished their parents had had more children so that they, the parents, would have had more help and thus easier lives, and that they, the respondents, would have more siblings to help them. Similarly, a majority of respondents wished they themselves would have more children, so as to have a better chance for more assistance now and in old age. It seems that these positive pressures outweigh the very real concerns for the costs, monetary as well as social, in raising and training children. I conclude that a pressure, summed up in the phrase "big family ideal," exists which helps to keep fertility fairly high in spite of individual predilections to the contrary. The predominance of responses for having a son as an heir indicates the importance of traditional Temne views. It may also provide a pressure for high fertility since in view of recognized high infant and child mortality, it is reasonable to conclude that having two or three sons ups the odds that one of them will survive. It is obvious from the data in Table 12 that the great majority of Temne, male and female, expect to be supported in old age/retirement by their child(ren), and that nearly as great a majority expect to live with one or another child. Once again the probability argument is applicable; the more children you have, the better they can support you. The "next generation," represented by the Magburaka secondary school series, seems to feel differently; this may merely reflect their younger ages or it may signal a shift in expectations regarding old age assistance and residence. It is evident that mean household size has decreased in recent years in rural areas and thus fewer Temne are raised in large households; 10quite possibly fewer of them

346

DORJAHN

want to live in retirement in a large household. Thus the pressure to have more children so as to maximize retirement possibilities may be easing, especially with the "next generation" and the urban segment.

Planning and Policy Matters Simon (1975:3) has argued that moderate (perhaps the key word) population growth has positive effects on the standard of living in the long run (after, say, 30 to 100 years) in both more-developed and less-developed countries-as compared to a stationary population and to very fast population growth. But of course any additional person adds a burden to parents and society in the short run. Therefore, whether one judges that population growth is good or bad-whether parents and society opt for more or less children-rationally depends upon how the importance of the long run is weighted relative to the short run.

Simon refers to the relative weighting of short-run and long-run futures as the discount rate; a high discount .rate means that one gives greater weight to the present and short-run future. Nearly all governmental regimes and politicians utilize high discount rates; they want demonstrable successes now for they may not be around to claim credit for long-run successes later, and the worsened short-run effects entailed in the quest of long-run gain can be politically deadly. Whatever the scientific merits of Simon's case-and I, for one, think they are impressive (see the papers in Section II, this volume)-I doubt they will be easy to sell to the political decision-making leadership of third world countries. Simon himself has cogently and honestly phrased the political stumbling block: "Of course nothing said here is intended to de-emphasize the short-run burden of additional children upon their families, and upon educational and other public facilities, until the system has time to adjust' (1975:11). Some African governments-today , the vast majority-may take no action; "socialist governments," following the Marxist lead, may conclude that however many children are born can be put to productive use in a socialist economy, and that efforts to reduce or halt population growth are unwarrented in what is essentially a "cornucopian situation." Ivory Coast, which has experienced faster economic than population growth, and Cameroon, plagued by unusually high rates of secondary sterility, have sought to promote fertility (Falkingham 1980), but such policies are likely to remain a tiny minority. I think most African governments which do formulate demographic policy will opt for fertility reduction.

TEMNE FERTILITY

347

The Government of Sierra Leone has been concerned with population growth and distribution both in terms of administration and of development, but no clear policy statements which would serve as a basis for action have been made (Dow and Benjamin 1975:454). Certain government officials have, however, indicated at various times and in different contexts that they favor lowered mortality, lowered fertility, and reduced or even reversed rural-urban migration. Unfortunately, these desires are not shared by all individual Temne, rural or urban, who contributed to this study. I would guess that these Temne share the goal of lowered mortality, do not necessarily share the wish for lowered fertility and do not share the desire to slow down or reverse ruralurban migration. Individuals would agree in the abstract with their government's wishes but would not necessarily act in accord with them themselves. Such discontinuities between the wishes of a government and the wishes of its people as individuals are not new. Neither are they limited to developing countries nor to issues regarding population. Most Sierra Leoneans would agree with their government leaders that the country needs technicians-diesel mechanics, agricultural extension agents and so on-and that the educational system should be geared to produce more people with such training. But for their own children, or for themselves, they prefer schools that prepare them for "white collar" jobs, preferably in government. Similarly, rising food prices and the disappointing trend of agricultural production in recent years have convinced most people that more labor (and government assistance) should be channelled into agriculture. Rural-urban migration reduces the rural labor force and most agree it is bad; but for themselves or their family members, the greater economic prospects in the towns should be and are seized. Thus, while most people see evidence that increasing population is a factor in the deteriorating man-land ratio and so on, and that lowered fertility may well be advantageous to the country, they are not all ready to forego the perceived individual advantages of larger families for themselves. It may be necessary to make sacrifices for the common good, but others should make them first. Discontinuities in the wishes of individuals and those of a government are obstacles to the success of government inspired and administered programs to achieve government policy goals. Such discontinuities, together with discontinuities between social science truth and politically-palatable truth, pose ethical problems for anthropologists or other social scientists wishing to contribute positively to the well-being of people in such high fertility countries as Sierra Leone. Should the anthropologist advocate what the perceived majority of individual Temne want, what the Government in power wants, or what social science analysis suggests as the wisest long-run

348

DORJAHN

policy? In some instances, it may be ethical to try to change policy-makers' positions; in others, to advise on policy-implementing programs to change the opinions of the people. In the latter instance, the question is one of how to do this. Advocates of government planning often times seem to assume that peoples' culturally molded attitudes can be nearly completely wiped away by governmental action; witness the thrust for a "new Soviet man," a "compleat ujamaa villager" and so on. It is not often so simple. How, then, can peoples' attitudes to fertility, in this context, be influenced, assuming that the Government of Sierra Leone should decide on such an action program? It seems to me that there are two possible strategies. First, people can be scared by graphic portrayals of the horrid prospects if they pursue their selfish, anti-social goal to have more children.1 1Food, land, and housing shortages, higher prices, a lowered level of living, increased social unrest · and so on are part of this Malthusian scenario. People will have fewer children when they are properly terrorized by the prospects of such a future. Second, people can be enticed or seduced by the glowing prospects of a coming "Golden Age" which looms reassuringly once population growth is slowed (or reversed). Having fewer children is a key step toward this better future characterized by improved amenities, a higher level of living, greater social integration and so on. Some parts of the economic foundation for this scenario are certainly suspect but there is a readily appreciated, simple logic to the argument that parents can provide more per capita help to two children than to six or twelve. Which strategy, the negative Malthusian or the positive Golden Age, is to be preferre~? Will either one or some mix of the two do the job? I suspect that a mix has the best chance with the Temne and, for that matter, with most peoples. I am fairly certain that a given mix will be more successful in one instance and less successful with another people. There will not be one, universally applicable, "correct" approach; rather the approach or strategy will be situation-specific with respect to different cultures, even in a comparatively small country such as Sierra Leone. Notes lJ am grateful to the National Science Foundation (Washington, D.C.) for making possible my research in Sierra Leone in 1954-55 and 1962-63, and to the National Institutes of Child Health and Human Development, Center for Population Research (Bethesda, Maryland) for supporting my 1976 research and that of Dr. Antony H. Smith. In section 3 I utilize some 1955 data; I collected no demographic data in Magburaka at that time.

TEMNE FERTILITY

2It

349

thus follows that demographic measures such as age-specific fertility rates, which depend upon data from registration systems, must either be approximated or omitted. 3Bogue and Palmore (1964:322) caution that: "If one area has higher child mortality rates than another, even though their fertility rates are the same, the area with the higher death rates will have a lower ratio." 4Smith (1981) has estimated the mean weaning age at 18 months for Magburaka; Dow (1971) has cited 16-17 months for provincial towns. 5Female single parent households comprised 5.5 percent of all households in 1963 (Dorjahn 1977:Table 3), and 15.7 percent of all households in 1976. 6Any neophyte who has chopped wood, spaded a garden, trundled dirt in a wheelbarrow, and so on, for an extended period knows that there are energy efficient and energy inefficient ways to perform each task. 7Mean household size in rural Mayoso has decreased from 8.2 in 1963 to 7.0 in 1976; while 16.2 percent of all households in 1963 were extended (contained two or more married men), only 10.2 percent were extended in 1976. 8Like all anthropologists I use "culture" and "cultural factors" as umbrella concepts which include the social, economic, political, educational, expressive, and so on, not as labels for a largely ignored residual category mentioned for the sake of completeness after the "important" factors-whether eeonomic, political, or what have you-have been indicated specifically. Like many, but not all anthropologists, I favor the view of culture as a system of shared ideas about how to behave/live. 9J'hree Mayoso men stated bluntly and categorically that a woman can't head a family. However, 5.5 percent of all households were headed by currently single mothers. 10J:n 1963, 15.6 percent of all people in Mayoso lived in small households (5 persons and less) and 33.0 percent in large households (15 persons and more); for 1976 the figUres were 21.7 percent in small and 24.1 percent in large households. 11 I am concerned here with possibly effective political strategies, not social science truth. At least in the short-run, population growth is likely to result in increased production per hectare through more labor intensive agriculture, but an effective political strategy can be based, at least in part, on untruths, part-truths, and so on. Professional social scientists may reject the Gotterdamerung Malthusian scenario; that doesn't mean that rural farmers reject it and that it can't be effective as an argument with such people for a government supported fertility limitation program.

References Cited Abelson, R. P. 1976 Social Psychology's Rational Man. In Rationality and the Social Sciences, S. I. Benn and G. W. Mortimore, eds. Pp. 58-89. London: Routledge and Kegan Paul. Abernethy, Virginia 1979 Population Pressure and Cultural Adjustment. New York: Human Sciences Press. Adams, William 1932 Ireland and Irish Emigration to the New World from 1815 to the Famine. New Haven: Yale University Press. Adu, S. V. 1969 Soils of the Navrongo-Bawku Area, Upper Region, Ghana. Soil Research Institute Memoirs Number 5. Kumasi, Ghana: Soil Research Institute. Ahn, Peter M. 1970 West African Soils. London: Oxford University Press. Aikin, John 1968 [orig. 1795] A Description of the Country from 30 to 40 Miles Round Manchester. Newton Abbot: David and Charles. Ainsworth-Davis, John 1924 Meat, Fish and Dairy Produce. The Resources of the Empire Series. London: Ernest Berm.

Alexander, Jennifer, and Paul Alexander 1978 Sugar, Rice and Irrigation in Colonial Java. Ethnohistory 25:207-23. 1979 Labour Demands and the 'Involution' of Javanese Agriculture. Social Analysis 3:22-44. 1982 'Shared Poverty' as Ideology: Agrarian Relationship in Nineteenth Century Java. Man 17:4: 597-619. Alexander, Paul 1982 Sri Lankan Fisherman: Rural Capitalism and Peasant Society. Canberra: Australian National University. Ames, Kenneth 1981 The Evolution of Social Ranking on the Northwest Coast of North America. American Antiquity 46:789-805. Anderson, Peter 1983 The Reproductive Role of the Human Breast. Current Anthropology 24:25-45. Anonymous 1847 Thoughts on Ireland. London: James Ridgway. Anonymous 1974 VII Censo de Poblaci6n. Guatemala: Dirreci6n General de Estadistica. Aono, T., et al. 1976 Impaired LH Release Following Exogenous Estrogen Administration in Patients with Amenorrhea-Galactorrhea Syndrom. Journal of Clinical Endocrinology and Metabolism 422:696-702.

350

351 Arensberg, Conrad, and Solon Kimball 1968 [orig. 1940) Family and Community in Ireland. Cambridge: Harvard University Press. Arnold, F., eta/. 1975 The Value of Children: A Cross-National Study, Vol. 1. Honolulu: East-West Population Institute. Barrera, Mario 1979 Race and Class in the Southwest: A Theory of Racial Inequality. Notre Dame, IN: Notre Dame Univ. Press. Barlett, Peggy 1980 The Anthropological Approach to Fertility Decision-Making. In Demographic Behavior: Interdisciplinary Perspectives on Decision-Making, T. K. Burch, ed. Boulder: Westview Press. Bawku District Mobile Health Service (BDMHS) (1975) Annual Report, 1974-75. Manuscript (in my files). Bean, F. D. 1975 Review of Economics of the Family. Demography 12:557-61. Beach, L. R., F. L. Campbell, and B. D. Townes 1979 Subjective Expected Utility and the Prediction of Birth Planning Decisions. Organizational Behavior and Human Performance 24:18-28. Beach, L. R., B. D. Townes, F. L. Campbell, and G. W. Keating 1976 Developing and Testing a Decision Aid for Birth Planning Decisions. Organizational Behavior and Human Performance 15:99-116. Becker, G. S. 1960 An Economic Analysis of Fertility. In Demographic and Economic Change in Developed Countries. National Bureau Committee for Economic Research, Conference Series 11. Pp. 209-31. Princeton: Princeton University Press. 1965 A Theory of the Allocation of Time. Economic Journal 75:493-517. 1976 The Economic Approach to Human Behavior. Chicago: University of Chicago Press. 1981 A Treatise on the Family. Cambridge, Mass: Harvard University Press. Becker, G. S., and H. G. Lewis 1973 Interaction Between Quantity and Quality of Children.In Economics of the Family: Marriage, Children, and Human Capital, T. W. Schultz, ed. Pp. 81-90. Chicago: University of Chicago Press for the National Bureau of Economic Research. Beer, T., eta/. 1979 Report of the Climatology, Agroclimatology and Water Resources Availability Sub Project. CSIR Northeast Ghana Savanna Project. Accra, Ghana: Water Resources Research Unit (C.S.I.R.). Bennett, John 1969 Northern Plainsmen. Chicago: Aldine Publishing Company. Berlin, Brent, Dennis E. Breedlove, and Peter H. Raven 1974 Principles of Tzeltal Plant Classification: An Introduction to the Botanical Ethnography of a Mayan-Speaking People of Highland Chiapas. New York: Academic Press. Breedlove, Dennis E. 1981 Flora of Chiapas. San Francisco: California Academy of Sciences. Besnard, Peter 1817 Minutes of the Trustees of the Linen and Hempen Manufactures of Ireland, Containing the Report of a Tour of Inspection Through the Provinces of Leinster, Munster and Connaught. Dublin: W. Folds and Sons.

352 Billewicz, W. Z., and I. A. McGregor 1981 The Demography of Two West African (Gambian) Villages, 1951-75. Journal of Biological Science 13:219-40. Birdwell-Pheasant, Donna 1984 Cycles of Powerlessness: Domestic Development and the "Labor-Flow" Enterprise. Paper presented to the 85th annual meeting of the American Anthropological Association, Denver. Bishop, Charles 1978 Culture and Biological Adaptations to Deprivation: The Northern Ojibwa Case. In Extinction and Survival in Human Populations, Charles Laughlin, Jr., and Ivan Brady, eds. Pp. 209-30. New York: Columbia University Press. Blackwood, Larry 1981 Alaska Native Fertility Trends, 1950-1978. Demography 18(2):173-79. Blake, J. 1968 Are Babies Consumer Durables? A Critique of the Economic Theory of Reproductive Motivation. Population Studies 22:5-25. Bledsoe, Caroline 1980 Women and Marriage in Kpelle Society. Stanford: Stanford University Press. 1981 The Manipulation of Kpelle Social Fatherhood. Ethnology 19:25-43. Bleek, W. 1976 Spacing of Children: Sexual Abstinence and Breastfeeding in Rural Ghana. Social Science and Medicine 10:225-30. Bleeker, P. 1849 Fragmenten eener reis over Java. Tijdschrift voor Nederlansch-Indie 2. 1863 Statistisch oekonomische onderzoekingen en beschouwingen op koloniaal gebied. Tijdschrift voor Nederlandsch-Indie (New Series) 1:198-99. Bodley, J. 1976 Anthropology and Contemporary Human Problems. Menlo Park: Cummings Publishing Co. Boeke, J. H. 1926 Inlandse Budgetten. Koloniale Studien X:272-321. 1948 Ontwikkelingsgang en Toekomst van Bevolkings- en Ondernemings- Landbouw in Netherlands-Indie. Leiden: E. J. Brill. Bogue, Donald J. 1969 Principles of Demography. New York: John Wiley and Sons, Inc. Bogue, Donald J., and J. A. Palmore 1964 Some Empirical and Analytic Relations Among Demographic Fertility Measures, with Regression Models for Fertility Estimation. Demography I:316-38. Bondestam, Lars 1980 The Political Ideology of Population Control. In Poverty and Population Control, Lars Bondestam and Staffan Bergstrom, eds. Pp. 1-38. New York: Academic Press. Bongaarts, John 1976 Intermediate Variables and Marital Fertility Rates. Population Studies 30:227:41. 1978 A Framework for Analyzing the Proximate Determinants of Fertility. Population and Development Review 4:105-32. 1980a Does Malnutrition Affect Fecundity? A Summary of Evidence. Science 208:564-69. 1980b The Fertility Inhibiting Effects of the Intermediate Fertility Variables. Paper prepared for the Seminar on the Analysis of Birth Histories; organized by the International Union for the Scientific Study of Population, London.

353 1982 The Fertility Inhibiting Effects of the Intermediate Fertility Variables. Studies in Family Planning 13:179-89. 1983 The Proximate Determinants of Natural Marital Fertility. In The Determinants of Fertility in Developing Countries. R.A. Bulatao and R.D. Lee, eds. Vol. 1. New York: Academic Press, pp. 103-138. Bongaarts, J., and H. Delgado 1979 Effects of Nutritional Status on Fertility in Rural Guatemala. In Natural fertility, H. Leridon and J. Menken, eds. Ordina: Liege. Bongaarts, J., and R. G. Potter 1982 Fertility, Biology and Behavior: An Analysis of the Proximate Determinants. New York: Academic Press. Boserup, E. 196S The Conditions of Agricultural Growth: The Economic of Agrarian Change Under Population Pressure. Chicago: Aldine. 1970 Women's Role in Economic Development. London: Allen and Unwin. Boulding, K. E. 1966 Economic Analysis, Fourth Edition - Volume 1: Microeconomics. New York: Harper and Row. Bracher, M.D., and Gigi Santow 1982 Breast-Feeding in Central Java. Population Studies 36:413430. Bradshaw, B. S., and F. D. Bean 1973 Trends in the Fertility of Mexican Americans: 1950-1970. Social Science Quarterly 53:688-96. Brainard, Jean 1980 Settlement and Fertility in a Nomadic Pastoral Population. Paper prepared for the 79th Annual Meeting of the American Anthropological Association, Washington, D.C. Brass, William, et a/. 1968 The Demography of Tropical Mrica. Princeton: Princeton University Press. Brintnall, Douglas 1979 Revolt Against the Dead. New York: Gordon and Breach. Bronte-Stewart, B., 0. Budtz-Olsen, J. Hickley, and J. Brock 1960 The Health and Nutritional Status of the Kung Bushmen of South West Africa. South Mrican Journal of Laboratory and Clinical Medicine 6:187-216. Brooks, G. 1975 Peanuts and Colonialism: Consequences of Commercialization of Peanuts in West Mrica, 1830-70. Journal of African History 16:22-54. Brown, J. K. 1970 A Note on the Division of Labor. American Anthropologist 72:1073-78. Bruger, D. H. 1975 Sociologische-Economische Geschiedenis van Indonesia, Vol. I. The Hague: Martinus Nijhoff. Bulatao, R. A. 1983 A Framework for the Study of Fertility Determinants. In The Determinants of Fertility in Developing Countries. R.A. Bulatao and R.D. Lee, eds. Vol. 1. New York: Academic Press, pp. 1-26. Bulatao, R.A., and F. Arnold 1977 Relationships Betweeen the Value and Cost of Children and Fertility: CrossCultural Evidence. Honolulu: East-West Population Institute. Bulatao, R.A., and R.D. Lee, eds. 1983 Determinants of Fertility in Developing Countries. 2 Volumes. New York: Academic Press.

354 Burley, David 1979 Cognitive Modelling: The Implications for Gold Rush Archaeology. Paper presented at the 1979 Canadian Archaeological Association Meetings, Vancouver, B.C. 1980 Marpole. Department of Archaeology, Simon Fraser University Publication No.8, Burnaby, B.C. Butz, William P. 1981 The Changing Role of Breastfeeding in Economic Development. Research in Human Capital and Development 2:95-117. 1982 Searching for a Natural Fertility Population. Discussion paper prepared for the UC-Berkeley Brown Bag Seminar in Demography, May 5, 1982. Cain, G. G., and M. D. Dooley 1976 Estimation of a Mod.el of Labor Supply. Fertility and Wages of Married Women. Journal of Political Economy 84 (Part II):S179-199. Cain, Mead T. 1977 The Economic Activities of Children in a Village in Bangladesh. Population and Development Review 3:201-27. Caldwell, J. C. 1967 Fertility Attitudes in Three Economically Contrasting Rural Regions of Ghana. Economic Development and Cultural Change 15:217-38. 1976 Toward a Restatement of Demographic Transition Theory. Population and Development Review 2:321-66. 1977a The Economic Rationality of High Fertility: An Investigation Illustrated with Nigerian Survey Data. Population Studies 31 :5-27. 1977b Towards a Restatement of Demographic Transition Theory. An Investigation of Conditions Before and at the Onset of Fertility Decline Employing Primarily African Experience and Data. In The Persistence of High Fertility, John C. Caldwell, ed. Pp. 25-119. Canberra: Australian National University Press. 1978 A Theory of Fertility: From High Plateau to Destabilization. Population and Development Review 4:553-77. 1980 Mass Education as a Determinant of the Timing of Fertility Decline. Population and Development Review 6:225-55. 1981 The Mechanisms of Demographic Change in Historical Perspective. Population Studies 35:5-29. 1982 Theory of Fertility Decline. New York: Academic Press. Caldwell, J. C., and B. Thompson 1975 Gambia. In Population Growth and Socioeconomic Change in West Africa, J. C. Caldwell, ed. Pp. 493-626. New York: Columbia Universiity Press. Caldwell, J. C., and P. Caldwell 1977 The Role of Marital Sexual Abstinence in Determining Fertility: A Study of the Yoruba in Nigeria. Population Studies 31:193-217. Caldwell, J.C., and Helen Ware 1977 The Evolution of Family Planning in an African City: Ibadan, Nigeria. Population Studies 33:487-507. Camarillo, Albert 1979 Chicanos in A Changing Society: From Mexican Pueblos to American Barrios in Santa Barbara and California, 1848-1930. Cambridge: Harvard University Press. Cantrelle, Pierre, and Henri Leridon 1971 Breast Feeding, Mortality in Childhood and Fertility in a Rural Zone of Senegal. Population Studies 25:505-23. Carney, F. J.

355 1975 Pre-Famine Irish Population: The Evidence from the Trinity College Estates. Irish Economic and Social History 2:3545. Carr-Saunders, A.M. 1936 World Population: Past Growth and Present Trends. Oxford: Clarendon Press. Carter, Jeanette E. 1970 Household Organization and the Money Economy in a Lorna Community, Liberia. Ann Arbor: University Microfilms. Catholic Communications Institute 1974 A National Study of Religious Attitudes. Dublin: Catholic Communications Institute. Chaudhury, R. H. 1977 Relative Income and Fertility. Demography 14:179-95. Chavez, A., and C. Martinez 1973 Nutrition and Development of Infants from Poor Rural Areas Ill. Nutrition Reports International 7:1-8. Chen, Lincoln 1983 Child Survival: Levels, Trends, and Determinants. In Determinants of Fertility in Developing Countries. R.A. Bulatao and R.D. Lee, eds. Vol. 1. New York: Academic Press, pp. 199-232. Chowdhury, A. 1978 Effect of Maternal Nutrition on Fertility in Rural Bangladesh. In Nutrition and Human Reproduction, W. Moseley, ed. Pp. 401-10. New York: Plenum Press. Clark, C., and M. R. Haswell 1970 The Economics of Subsistance Agriculture. New York: St. Martin's Press. Cleveland, David A. 1980a The Population Dynamics of Subsistence Agriculture in the West African Savanna: A Village in Northeast Ghana. Unpublished doctoral dissertation, University of Arizona. 1980b Demographic and Agricultural Change in Northeast Ghana. Paper presented at the Annual Meeting of the American Anthropological Association, Washington, D.C., December. Coale, Ansley J. 1974 The History of the Human Population. Scientific American 231:41-51. 1975 The Demographic Transition. In The Population Debate: Dimensions and Perspectives. Papers of the World Population Conference, Bucharest, 1974. Vol. 1. New York: United Nations. Coale, Ansley J., and T. James Trussell 1974 Model Fertility Schedules: Variations in the Age Structure of Childbearing in Human Populations. Population Index 40:185-258. 1975 Erratum. Population Index 41:572-3. 1978 Technical Note: Finding the Two Parameters that Specify a Model Schedule of Human Fertility. Population Index 44:203-11. Cochrane, Susan 1979 Fertility and Education: What Do We Really Know? World Bank Staff Occasional Papers, No. 26. Baltimore: Johns Hopkins Univ. Press. Cohen, J. 8., L. J. Severy, and 0. T. Ahtola 1978 An Extended Expectancy-Value Approach to Contraceptive Alternatives. Journal of Population 1:2241. Cohen, Mark 1985 Prehistoric Hunter-Gatherers: The Meaning of Social Complexity. In Prehistoric

356 Hunter-Gatherers, T. D. Price and James Brown, eds. Pp. 99-119. New York: Academic Press. Collier, George 1975 Fields of The Tzotzil. Austin: University of Texas Press. Colvin, L. B 1981 The Gambia. In The Uprooted of the Western Sahel: Migrants' Quest for Cast in Senegambia, L. G. Colvin, et al., eds. Pp. 387-413. New York: Preager. Commission on the Status of Women 1972 Report to the Minister of Finance. Dublin: Stationery Office. Connell, K. H. 1965 Land and Population in Ireland, 1780-1845. In Population in History: Essays in Historical Demography,. D. V. Glass and D. E. C. Eversley, eds. Pp. 423-33. London: Edward Arnold. Cowgill, George L. 1975 On Causes and Consequences of Ancient and Modern Population Changes. American Anthropologist 77:505-25. Crimmins-Gardner, E., and P. A. Ewer 1978 Relative Status and Fertility. In Research in Population Economics, Vol. 1, J. L. Simon, ed. Pp. 247-59. Greenwich, Connecticut: JAI Press. Crosbie, P. V., and D. Bitte 1982 A Test of Luker's Theory of Contraceptive Risk-Taking. Studies in Family Planning 13:67-78 Currens, Gerald E. 1974 The Lorna Farmer: A Socioeconomic Study of Rice Cultivation and the Use ·of Resources among a People of Northwestern Liberia. Ann Arbor: University Microfilms. Curtin, P. D. 1975 Economic Change in Pre-Colonial Africa: Senegambia in the Era of the Slave Trade. Madison: University of Wisconsin Press. Dahlberg, Frances 1976 More on Mechanisms of Population Growth. Current Anthropology 17:164-66. Dale, E., D. Gerlach, and A. Wilhite 1979 Menstral Dysfunction in Distance Runners. Ostetrics and Gynecology 54:47-53. DaVanzo, Julie, W.P. Butz, and J.-P. Habicht 1983 How Biological and Behavioral Influences on Mortality in Malaysia Vary During the First Year of Life. Population Studies 37:381-402. Davidson, A. R., imd J. J. Jaccard 1975 Population Psychology: A New Look at an Old Problem. Journal of Personality and Social Psychology 31:1073-82. 1976 Social-Psychological Determinants of Fertility Intentions. In Population Educational Issues, S. H. Newman and V. D. Thompson, eds. Pp. 131-37. Washington, D.C.: U.S. Government Printing Office. 1979 Variables that Moderate the Attitude-Behavior Relation: Results of a Longitudinal Survey. Journal of Personality and Social Psychology 37:1364-76. Davis, Kingsley 1963 The Theory of Change and Response in Modern Demographic History. Population · Index 29:345-66. 1977 Population Policy and the Theory of Reproductive Motivation. Economic Development and Cultural Change 25 (Supplement): 159-79.

357 Davis, Kingsley and Judith Blake 1956 Social Structure and Fertility: An Analytic Framework. Economic Development and Cultural Change 4:211-35. Davis, T. L. F. 1973 Fertility Differentials Among the Tribal Groups of Sierra Leone. Population Studies 27:501-14. Day, William 1862 The Famine in the West. Dublin: Hodges, Smith and Co. Deevey, Edward, Jr. 1968 Discussion. In Lee and Devore. P. 286. DeLamater, J. and P. MacCorquodale 1979 Premarital Sexuality. Madison: University of Wisconsin Press. Delgado, H., et a/. 1978 Nutrition and Birth Interval Components: The Guatemala Experience. In Nutrition and Human Reproduction, W. Moseley, ed. Pp. 385-99. New York: Plenum Press. 1979 Effect of Maternal Nutritional Status and Infant Supplementation During Lactation on Postpartum Amenorrhea. American Journal of Obstetrics and Gynecology 135:303-7. Delvoye, P., eta/. 1977 The Influence of the Frequency of Nursing and of Previous Lactation Experience on Serum Prolactin on Nursing Mothers. Journal of Biosocial Science 9:447-51. 1978 Serum Prolactin, Gonadotrophins, and Estradiol in Menstruating and Amenorrhic Mothers During Two Years Lactation. American Journal of Obstetrics and Gynecology 139:635-39. Department of Agriculture and Technical Instruction for Ireland 1902 Ireland: Industry and Agriculture. Dublin: Browne and Nolan. DeTray, D. N. 1973 Child Quality and the Demand for Children./n Economics of the Family: Marriage, Children and Human Capital, T. W. Schultz, ed. Pp. 81-90. Chicago: University of Chicago Press, for the National Bureau of Economic Research. · Diamond, M., P. G. Steinhoff, J. A. Palmore and R. G. Smith 1973 Sexuality, Birth Control and Abortion: A Decision-Making Sequence. Journal of Biosocial Science 5:347-61. Dickemann, M. 1975 Demographic Consequences of Infanticide in Man. Annual Review of Ecology and Systematics 6:107-37. Diop, M. 1972 Histoire des Oasses Social dans 1'Afrique de l'Ouest: le Senegal. Paris: Maspero. Dirks, Robert 1978 Resource Fluctuations and Competitive Transformations in West Indian Slave Societies. In Extinction and Survival in Human Populations. Charles Laughlin, Jr. and Ivan Brady, eds. Pp. 122-180. New York: Columbia University Press. 1980 Social Responses During Severe Food Shortases and Famine. Current Anthhropology 21:21-44. Dodd, George 1856 The Food of London. London: Lonaman, Brown, Green and Longmans. Donnelly, James 1975 The Land and the People of Nineteenth Century Cork. London: Routledge and Kegan Paul.

358 Dorjahn, Vernon R. 1958 Fertility, Polygyny and Their Interrelations in Temne Society. American Anthropologist 60:838-60. 1975 Migration in Central Sierra Leone: The Temne Chiefdom of Kolifa Mayoso. Africa 45:29-49. 1976 Rural-Urban Differences in Infant and Child Mortality Among the Temne of Kolifa. Journal of Anthropological Research 32:74-103. 1977 Temne Household Size and Composition: Rural Changes Over Time and RuralUrban Differences. Ethnology 16:105-27. m.s. Migration and the Growth of Magburaka, a Sierra Leone City. Manuscript. University of Oregon, Eugene. Dorjahn, Vernon R., and Barry L. Isaac, eds. 1979 Studies in the Economic Anthropology of Liberia and Sierra Leone. Liberian Studies Monograph No. 6. Philadelphia: Institute for Liberian Studies. Dow, T. E., Jr. 1971 Fertility and Family Planning in Sierra Leone. Studies in Family Planning 2:153-65. Dow, Thomas E., and E. Benjamin 1975 Demographic Trends and Implications. In Population Growth and Socioeconomic Change in West Africa. Caldwell, J. C., et al., eds. New York: Columbia University Press. Drake, Michael 1963 Marriage and Population Growth in Ireland, 1750-1845. Economic History Review 16:307-13. 1969 Population Growth and the Irish Economy. In The Formation of the Irish Economy, L. M. Culllen, ed. Pp. 65-76. Cork: The Mercier Press. Dublin Mansion House Relief Committee 1881 The Irish Crisis of 1879-80. Dublin: Browne and Nolan. Dusenberry, J. S. 1960 Comment. In Demographic and Economic Change in Developed Countries. Pp. 231-37. Princeton: Princeton University Press. Dumbuya, M. B. 1963 Social Effects of Industrialization and Urbanization in Sierra Leone. Africana Research Bulletin 3:54-72. Dumond, Don E. 1975 The Limitation of Human Population: A Natural History. Science 187:713-21. Dumsmore, J., et al. 1976 The Agricultural Development of the Gambia: An Agricultural, Environmental and Socioeconomic Analysis. Surbitron: Land Resources Division, Ministry of Overseas Development. Dunn, J., B. Nisula and D. Rodbard 1981 Transport of Steroid Hormones: Binding of 21 Endogenous Steroids to Both Testosterone-Binding Globulin and Corticosteroid-Binding Globulin in Human Plasma. Journal of Clincial Endocrinology and Metabolism 53:58-68. Dunsmore, J., et al. 1976 The Agricultural Development of the Gambia: An Agricultural, Environmental and Socioeconomic Analysis. Surbiton: Land Resources Division, Ministry of Overseas Development. Dutton, Hely 1824 A Statistical and Agricultural Survey of the County of Galway. Dublin: Dublin University Press.

359 Dzegede, S. A. 1981 Urbanization and Fertility Decline in West Africa: Ghana, Sierrra Leone, and Liberia. Journal of Comparative Family Studies 12:233-244. Easterlin, Richard A. 1966 On the Relation of Economic Factors to Recent and Projected Fertility Changes. Demography 3:131-53. 1968 Population, Labor Force, and Long Swings in Economic Growth: The American Experience. New York: National Bureau of Economic Research. 1969 Toward a Socio-Economic Theory of Fertility. In Fertility and Family Planning: A World View, L. Corsa, Jr., and R. Freedman, eds. Pp. 127-56. Ann Arbor: University of Michigan Press. 1973 Relative Economic Status and the American Fertility Swing. In Family Economic Behavior: Problems and Prospects, E. B. Sheldon, ed. Pp. 170-223. Philadelphia: Lippincott. 1975 An Economic Framework for Fertility Analysis. Studies in Family Planning 6:54-63. 1978 The Economics and Sociology of Fertility: A Synthesis. In Historical Studies of Changing Fertility, C. Tilly, ed. Pp. 57-133. Princeton: Princeton University Press. 1983 Modernization and Fertility. In Determinants of Fertility in Developing Countries. R.A. Bulatao and R.D. Lee, eds. Vol. 2. New York: Academic Press, pp. 562-86. Easterlin, R. A., R. A. Pollak, and M. L. Wachter 1980 Toward a More General Economic Model of Fertility Preferences. In Population and Economic Change in Developing Countries, R. A. Easterlin, ed. Pp. 81-149. Chicago: University of Chicago Press. Eaton, Joseph and A. J. Mayer 1954 Man's Capacity to Reproduce: The Demography of a Unique Population. Glencoe, Illinois: The Free Press. Edwards, R. Dudley and T. Desmond Williams (eds.) 1956 The Great Famine: Studies in Irish History, 1845-52. Dublin: Irish Committee of Historical Sciences. Elly, Sandham 1848 Potatoes, Pigs, and Politics: The Curse of Ireland, and the Cause of England's Embarrassments. London: Kent and Richards. Elson, R. E. 1978 The Impact of Government Sugar Cultivation in the Pasuruan Area, East Java, Durina the Cultivation System Period. Review of Indonesian and Malayan Stuudies 12:26-55. Brmisch, J. 1979 The Relevance of the 'Easterlin Hypothesis' and the 'New Homme Economics' to Fertility Movements in Great Britian. Population Studies 33:39-58. Fage, John D. 1964 Reflections on the Early History of the Mossi-Dagomba Group of States. In The Historian in Tropical Africa. J. Vansina, R. Mauny, and L. V. Thomas, eds. Pp. 177-91. London: Oxford University Press. Family Plannins Perspectives Volume 14, 112, March-April, p. 107. Faruqee, Rashid 1979 Sources of Fertility Decline: Factor Analysis of Inter-Country Data. World Bank Staff Working Paper No. 318. Fasseur, C. 1975 Kulturstelsel en Koloniale Baten. Leiden: Universitaire Pers.

360 Faulkingham, Ralph H. 1977 Fertility in Tudu: An Analysis of Constraints on Fertility in a Village in Niger. In The Persistence of High Fertility, John C. Caldwell, ed. Pp. 153-88. Canberra, Australia: Australian National University. 1980 West African Fertility: Levels, Trends, and Determinants. Report submitted to REDSO/WA, U.S.A.I.D., Abidjan, Ivory Coast. Fawcett, J. T. and F. Arnold 1973 The Value of Children: Theory and Method. Representative Research in Social Psychology 4:23-35. Fishbein, M., J. J. Jaccard, A. R. Davidson, I. Ajzen, and B. Loken 1980 Predicting and Understanding Family Planning Behaviors. In Understanding Attitudes and Predicting Social Behavior, I. Ajzen and M. Fishbein, eds. Pp. 130-47. Englewood Cliffs, N.J.: Prentice-Hall. Flanagan, Donal 1975 The More Subtle Discrimination. Studies: An Irish Quarterly Review of Letters, Philosophy, and Science. Forney, et a/. 1981 Aromatization of Androstenedione to Estrone by Human Adipose Tissue in Vitro. Correlation with Adipose Tissue Mass, Age, and Endometrial Neoplasia. Journal of Clinical Endocrinology and Metabolism 53:192-99. Fortes, Meyer 1945 The Dynamics of Clanship Among the Tallensi. London: Oxford University Press. 1949 The Web of Kinship Among the Tallensi. London: Oxford University Press. 1953 The Structure of Unilineal Descent Groups. American Anthropologist 55:17-41. Foster, Thomas Campbell 1846 Letters on the Condition of the People of Ireland. London: Chapman and Hall. Fox, G.L. 1980 The Mother-Adolescent Daughter Relationship as a Sexual Socialization Structure: A Research Review. Family Relations 29:21-28. Francis, W. J. 1970 Reproduction at Menarche and Menopause in Women. Journal of Reproduction and Fertility, suppl. 12:89-98. Frank, Andre Gunder 1967 Sociology of Development and Underdevelopment of Sociology. Catalyst 3:20-73. Freedman, R., A. I. Hermalin and M. C. Chang 1975 Do Statements About Desired Family Size Predict Fertility? The Case of Taiwan, 1967-1970. Demography 12:407-16. Freeman, T. W. 1957 Pre-Famine Ireland: A Study in Historical Geography. Manchester: Manchester University Press. Fried, E. S., S. L. Hofferth and J. R. Udry 1980 Parity-Specific and Two-Sex Utility Models of Reproductive Intentions. Demography 17: 1-11. Fried, E. S., and J. R. Udry 1979 Wives' and Husbands' Expected Costs and Benefits of Childbearing as Predictors of Pregnancy. Social Biology 26:265-73. Friedl, John, and Walter Ellis 1976 Celibacy, Late Marriage and Potential Mates in a Swiss Isolate. In Anthropological Studies of Human Fertility, Bernice Kaplan, ed. Pp. 23-35. Detroit: Wayne State University Press.

361 Frisch, R. E. 1972 Weight at Menarche: Similarity for Well-nourished and Undernourished Girls at Different Ages, and Evidence for Historical Constancy. Pediatrics 50:445-50. 1975 Demographic Implications of the Biological Determinants of Female Fecundity. Social Biology 22:17-22. 1978a Nutrition, Fatness, and Fertility: The Effect of Food Intake on Reproductive Ability. In Nutrition and Human Reproduction, W. Moseley, ed. Pp. 91-122. New York: Plenum Press. 1978b Population, Food Intake and Fertility. Science 199:22-30. 1980 Pubertal Adipose Tissue: Is it Necessary for Normal Sexual Maturation? Evidence from the Rat and Human Female. Federation Proceedings 39:2395-2400. 1982 Malnutrition and Fertility. Science 215:1272-73. Frisch, R. E., and J. W. MacArthur 1974 Menstrual Cycles: Fatness as a Determinant of Minimum Height for Weight Necessary for their Maintenance or Onset. Science 185:949-51. Frisch, R., and R. Revelle 1970 Height and Weight at Menarche and a Hypothesis of Critical Body Weights and Adolescent Events. Science 169:397-98. Frisch, R., R. Revelle, and S. Cook 1971 Reply to Johnston, Malina and Galbraith. Science 174:1148-49. Frisch, R., J. Canick, and D. Tulchinsky 1980 Human Fatty Marrow Aromatizes Androgen to Estrogen. Journal of Clincial Endocrinology and Metabolism 51 :394-96. Fryer, D. W., and J. C. Jackson 1977 Indonesia. London: Ernest Benn. Furnivall, J. S. 1944 Netherlands India: A Study of Plural Economy. Cambridge University Press. Furstenberg, G. · 1972 Unplanned Parenthood: The Consequences of Teenage Childbearing. New York: Free Press. Gaisie, S. K. 1976 Estimating Ghanaian Fertility, Mortality, and Age Structure. University of Ghana Population Studies, No. 5. Legon, Ghana: The Population Dynamics Program, University of Ghana. Gaisie, S. K., N. 0. Addo, and S. B. Jones 1975 Population Policy and its Implementation. In Population Growth and Socioeconomic Change in West Africa, John C. Caldwell, ed. Pp. 408-24. New York: Columbia University Press. Gaisie, S. K., and A. S. David 1974 Planned Fertility Reduction in Ghana: Structural Interrelationships, Potential Socio-economic Impact, and the Magnitude of the Needed Programs. Thana Population Series No. 6. Legon, Ghana: Population Dynamics Program, University of Ghana. Gambia CDSS FY 82: Small Program Statement 1982 Washington: USAID. Gardner, B. 1973 Economics of the Size of North Caroline Rural Families. In Economics of the Family: Marriage, Children, and Human Capital, T. W. Schultz, ed. Pp. 160-63. Chicago: University of Chicago Press, for the National Bureau of Economic Research. Geertz, c. 1963 Agricultural Involution: The Process of Ecological Change in Indonesia. Berkeley and Los Angeles: University of California Press.

362 1973 Comments on Benjamin White's 'Demand for Labour and Population Growth in Colonia Java.' Human Ecology 1:237-39. Gelpke, J. H. F. S. 1878 Rapport over de Padi-Cultuur in de Afdeeling Ngrowo. Batavia: Landscrukkerij. Ghana Census Office (GCO) 1964 Demographic Characteristics of Local Authorities, Regions and Total Country. 1960 Population census of Ghana, Volume 3. Accra: Ministry of Broadcasting and Information on behalf of the Central Bureau of Statistics. 1975 Demographic Characteristics of Local Authorities, Regiona and Total Country. 1970 Population Census of Ghana, Volume 3. Accra: Ministry of Broadcasting and Information on behalf of the Central Bureau of Statistics. 1975 Demographic Characteristics of Local Authorities, Regiona and Total Country. 1970 Population Census of Ghana, Volume 3. Accra: Ministry of Broadcasting and Information on behalf of the Central Bureau of Statistics.) Ghana National Archives (GNA) n.d. [c. 1911] Census taking 1911. ADM 56/1/109. n.d. [c. 1921) Census taking 1921. ADM 56/1/241, case number 52/1919. n.d. [c. 1931a) Census report on the Kusasi District, excluding the mandated area of Togoland. ADM 57/5/8. (From the District Commissioner's Office, Bawku.) n.d. [c. 1931b] Census in the mandated area of Togoland which is incorporated in the Kusasi District. ADM 57/5/8. (From the District Commissioner's Office, Bawkuu.) Gibbs, James L., Jr. 1965 The Kpelle of Liberia. In Peoples of Africa, James L. Gibbs, Jr., ed. Pp. 197-240. New York: Holt, Rinehart and Winston. Giddens, Anthony 1976 Classical Social Theory and the Origins of Sociology. American Journal of Sociology 81:703-29. Gill, Conrad 1925 The Rise of the Irish Linen Industry. Oxford: Clarendon Press. Gille, H. and R. H. Pardoko 1966 A Family Life Study in East Java. In Family Planning and Population Programmes, B. Berelson, et at., eds. Chicago: University of Chicago Press. Gillmor, Desmond 1977 Agriculture in the Republic of Ireland. Budapest: Akademiai Kiado. Glick, Paul C. 1977 Updating the Life Cycle of the Family. Journal of Marriage and the Family 39:5-13. Glick, Paul C., and Robert Parke, Jr. 1965 New Approaches to Studying the Life Cycle of the Family. Demography 2:187-202. Gonzalez, N. 1964 Lactation and Pregnancy: A Hypothesis. American Anthropologist 68:873-78. Goode, W. J. 1982 The Family, 2nd ed. Englewood Cliffs, N.J.: Prentice-Hall. Gordon, Gillian Mary 1973 An Evaluation of Nutrition Education Given to Pregnant and Nursing Women in Six Communities in the Upper Region of Ghana. M.Sc. Thesis, Department of Nutrition and Food Science, Faculty of Science, University of Ghana. Gould, J.P., and C. F. Ferguson 1980 Microeconomic Theory. Homewood, Ill.: Richard D. Irwin. Government of the Gold Coast (GOGC)

363 Government of the Gold Coast (GOGC) 1923 1921 Population Census of the Gold Coast Report. Accra: Crown Agents for the Colonies. 1932 1931 Census of the Gold Coast Report. Accra: Crown Agents for the Colonies. 1950 1948 Population Census of the Gold Coast Report. London: Crown Agents for the Colonies. Graaf, H. de 1949 Geschiedenis van Indonesi~. The Hague and Bandung: W. van Hoeve. Graff, Harvey · 1979 Literacy, Education and Fertility, Past and Present: A Critical Review. Population and Development Review 5:105-40. Graves, N. B., and T. D. Graves 1978 The Impact of Modernization on the Personality of a Polynesian People, or, How to Make an Up-tight, Rivalrous Westerner out of an Easy-going, Generous Pacific Islander. Human Organization 37:115-135. Gray, J. M. 1966 [1940] A History of The Gambia. London: Frank Cass. Gray, R. H. 1976 The Menopause: Epidemiological and Demographic Considerations. In The Menopause: A Guide to Current Research and Practice, R. J. Beard, ed. Pp. 25-40. Lancaster, England: MTP Press. 1982 The Impact of Health and Nutrition on Natural Fertility. In Determinants of Fertility in Developing Countries: A Summary of Knowledge, R. A. Bulatao and R. D. Lee with the assistance of P. E. Hollerbach and J. Bongaarts, eds. Panel on Fertility Determinants, Report No.3. Washington, D. C.: National Academy Press. Gray, W. C. and Staff Surgeon Dochard 1825 Travels in Western Africa in the Years 1818-1821. London: J. Murray. Green, E. R. R. 1956 Agriculture. In The Great Famine: Studies in Irish History, 1845-52, R. Dudley Edwards and T. Desmond Williams, eds. Pp. 89-128. Dublin: Irish Committee of Historical Sciences. Grodin, J., P. Siiteri, and P. MacDonald 1973 Source of Estrogen Production in Postmenopausal Women. Journal of Clinical Endocrinology and Metabolism 36:207-14. Grosser, E., editor 1975 Problematica y Perspectivas de Desarrollo en Ia Agricultura Migratoria de una Subregion de la Selva Lacandona. Berlin: Institut fuP Sozialokonomie der Agraentwicklung Technische Universitiit Berlin. Gurin, P., G. Gurin, and B. M. Morrison 1978 Personal and Ideological Aspects of Internal and External Control. Social Psychology 41:275-96. Gwatkin, Davidson R. 1975 Governmental Population Policies. In Population Growth and Socioeconomic Change in West Africa, John C. Caldwell, ed. Pp. 169-84. New York: Columbiia University Press. Hajnal, J. 1965 European Marriage Patterns in Perspective. In Population and History: Essays in Historical Demography, D. C. Glass and D. E. C. Eversley, eds. Pp. 101-43. Chicago: Aldine Publishing Company. Hall, C. J. J. van, and C. van de Koppel 1946-8 De Landbouw in de Indische Archipel, 4 Vols. The Haaue: W. van Hoeve.

364 Hall, Mr. and Mrs. S. C. 1841-43 Ireland: Its Scenary, Character, Etc. London: Jeremiah How. 3 Vol. Handwerker, W. Penn 1973 Technology and Household Configuration in Urban Africa: The Bassa of Monrovia. American Sociological Review 38:182-97. 1977 Family, Fertility, and Economics. Current Anthropology 18:259-87. 1978 Viability, Location, and Timing of Liberian Periodic Markets. In Marketplace Trade-Periodic Markets, Hawkers, and Traders in Africa, Asia, and Latin America. Robert H. T. Smith, ed. Pp. 119-221. Vancouver: Centre for Transportation Studies, University of British Columbia. 1981a Reproductive Choices and Behavior: A Test of Two Theories of Fertility Variation with Data from Monrov.ia, Liberia. Medical Anthropology 5:261-91. 1981b Westernization, Family Moralities, and Economic Constraints: Response to Caldwell. Medical Anthropology 5:368-81. 1981c Productivity, Marketing Efficiency, and Price- Support Programs: Alternative Paths to Rural Development in Liberia. Human Organization 40:27-39. 1981d African Marketing Institutions and Rural Development Policy. African Urban Notes 10:5-20. 1983 The First Demographic Transition: An Analysis of Subsistence Choices and Reproductive Consequences. American Anthropologist 85:5-27. 1984 Parameter Estimation in Microdemography: Strategies for Age Specific Fertility Ratios. Paper presented to the 82nd Annual Meeting of the American Anthropological Association, Chicago. 1986 The Modern Demographic Transition: An Analysis of Subsistence Choices and Reproductive Consequences. American Anthropologist 88: in press. 1987 Fiscal Corruption and the Moral Economy of Resource Acquisition. Studies in Economic Anthropology 9: in press. Handwerker, W.P., and P.V. Crosbie 1982 Sex and Dominance. American Anthropologist 84:97-104. 1983 On Methodology and Strawrnen: Reply to Turke and Betzig. American Anthropologist 85:411. Harpending, H. 1971 !Kung Hunter-Gatherer Population Structure. Unpublished doctoral dissertation, Harvard University. Harpending, Henry, and LuAnn Wandsnider 1981 Population Structures of Ghanzi and Ngamiland !Kung./n Current Developments in Anthropological Genetics, Michael Crawford and James Mielke, eds. New York: Plenum. Harrell, B. B. 1981 Lactation and Menstruation in Cultural Perspective. American Anthropologist 83:796-823. Harris, Marvin 1971 Culture, Man & Nature. New York: Crowell. Hart, K. 1982 The Political Economy of West African Agriculture. New York: Cambridge University Press. Haswell, M. R. 1963 The Changing Pattern of Economic Activity in a Gambia Village. London: HMSO. 1975 The Nature of Poverty. New York: St. Martin's Press. 1981 Food Consumption in Relation to Labor Output. In Seasonal Dimensions to Rural

365 Poverty, R. Chambers, eta/., eds. Pp. 38-41. Totawa, New Jersey: Allanheld, Osmun and Company. Hauser, R. M., and D. L. Featherman 1977 The Process of Stratification. New York: Academic Press. Hawthorn, Geoffrey 1970 The Sociology of Fertility. London: Collier-Macmillan, Ltd. 1978 Introduction. Journal of Development Studies 14:1-21. Hayden, Brian 1972 Population Control Among Hunter/Gathers. World Archaeology 4:205-21. 1981a Subsistence and Ecological Adaptations of Modem Hunter/Gatherers. In Omnivorous Primates: Gathering and Hunting in Human Evolution, Robert S. 0. Harding and Geza Teleki, eds. New York: Columbia University Press. 1981b Research and Development in the Stone Age: Technological Transitions Among Hunter-Gatherers. Current Anthropology 22:519-48. Hayden, Brian, and Aubrey Cannon 1982 The Corporate Group as an Archaeological Unit. Journal of Anthropological Archaeology I:132-58. Hechter, Michael 1975 Internal Colonialism: The Celtic Fringe in British National Development, 1536-1966. Berkeley: University of California Press. Heer, David M. 1983 Infant and Child Mortality and the Demand for Children. In Determinants of Fertility in Developing Countries. R.A. Bulatao and R.D. Lee, eds. Vol. 1. New York: Academic Press. pp. 369-87. Henry, Louis 1961 Some Data on Natural Fertility. Eugenics Quarterly 8:81-91. 1965 French Statistical Research. In Public Health and Population Change, M. C. Sheps and J. C. Ridley, eds. Pp. 333-50. Pittsburgh: University of Pittsburgh Press. 1979 Concepts Actuels et Resultats Empiriques sur Ia Fecondite Naturalle. In Natural Fertility, Henri Leridon and Jane Menken, eds. Liege: Ordina. Hilton, Thomas Eric 1962 Notes on the History of Kusasi. Transactions of the Historical Society of Ghana 6:79-96. Ho, T. J. 1979 Time Costs of Child Rearing in the Rural Philippines. Population and Development Review 5:643-62. Hoffman, L. W., and M. L. Hoffman 1973 The Value of Children to Parents. In Psychological Perspectives on Population, J. T. Fawcett, ed. Pp. 19-76. New York: Basic Books. Hoffman, L. W., and J.D. Manis 1979 The Value of Children in the United States. Journal of Marriage and the Family 41:583-96. Howell, Nancy 1976 The Population of the Dobe Area !Kung. In Kalahari Hunter-Gatherers. Studies of the !Kung San and Their Neighbors, Richard B. Lee and Irven DeVore, eds. Pp. 137-51. Cambridge, Massachusetts: Harvard University Press. 1979 Demography of the Dobe !Kung. New York: Academic Press. Huffman, Sandra L., et a/. 1978a Nutrition and Post-partum Ammenorrhoea in Rural Bangladesh. Population Studies 32(2):251-60.

366 1978b Postpartum Amenorrhea: How is it Affected by Maternal Nutritional Status? Science 200:1155-7. 1980 Lactation and Fertility in Rural Bangladesh. Population Studies 34:337-47. Hull, T. H., and V. J. 1976 Social And Economic Support for High Fertility in Indonesia. Yogyakarta: Lembaga Kependudukan. Hull, V. J. 1976 The Positive Relation Between Economic Class and Family Size. Yogyakarta: Lembaga Kependudukan. 1977 A Study of Birth Interval Dynamics in Rural Java. In Nutrition and Human Reproduction, W. H. Mosely, ed. New York: Plenum Press. Hunn, Eugene S. 1977 Tzeltal Folk Zoology: The Classification of Discontinuities in Nature. New York: Academic Press. Hunter, John M. 1968 The Clans of Nangodi. Africa 38(4):377-412. Russ-Ashmore, R. 1980 Fat and Fertility: Demographic Implications of Differential Fat Storage. Yearbook of Physical Anthropology 23:65-91. Hutchinson, G. Evelyn 1978 An Introduction to Population Ecology. New Haven, Connecticut: Yale University Press.

IRDC

1980 Nutritional Status of the Rural Population of the Sahel. Ottowa: International Development Research Center. Irons, William 1977 Evolutionary Biology and Human Fertility. Paper presented at the Annual Meeting of the American AnthropOlogical Association, Houston, TX. Isaac, Barry L. 1979 Female Fertility and Marital Form Among the Mende of Rural Upper Bambara Chiefdom, Sierra Leone. Ethnology 19:297-313. Jaccard, J. J .., and A. R. Davidson 1972 Toward an Understanding of Family Planning Behaviors. Journal of Applied Social Psychology 2:228-35. 1976 The Relation of Psychological, Social, and Economic Variables to Fertility-Related Decisions. Demoography 13:329-38. Johnson, J. H. 1970 The Two "lrelands" at the Beginning of the Nineteenth Century. In Irish Geographical Studies, in Honour of E. Estyn Evans. Nicholas Stephens and Robin Glasscock, eds. Pp. 224-43. Belfast: The Queen's University of Belfast. Johnston, F., R. Malina, and M. Galbarth 1971 Height, Weight, and Age at Menarche and the "Critical Weight" Hypothesis. Science 174:1148. Johnston, F., A. Roche, L. Schell, and H. Wettenhall 1975 Critical Weight at Menarche: Critique of a Hypothesis. American Journal of Diseases of Children 129:19-23. Kennedy, Robert 1973 The Irish: Emigration, Marriage and Fertility. Berkeley: University of California Press.

367 Kerr, Barbara 1943 Irish Seasonal Migration to Great Britain, 1800-38. Irish Historical Studies 4:365-80. Keys, A. J., et at. 1950 The Biology of Human Starvation. Minneapolis: University of Minnesota. Kippley, S., and J. Kippley 1974 The Relation Between Breastfeeding and Amenorrhea: Report of a Survey. In Breastfeeding and Natural Child Spacing, S. Kippley, ed. Pp. 163-68. Klibanski, A., et at. 1981 Reproductive Function During Fasting in Men. Journal of Clinical Endocrinology and Metabolism 53:258-63. Knight, G. R. 1980 From Plantation to Padi Field. Modern Asian Studies 14:177-204. Knodel, John 1977a Family Limitation and the Fertility Transition: Evidence from the Age Patterns of Fertility in Europe and Asia. Population Studies 31:219-49. 1977b Breastfeeding and Population Growth. Science 198:1ll-15. 1979 From Natural Fertility to Fertility Limitation: The Onset of Fertility Transition in a Sample of German Villages. Demography 16:493-521. 1983 Natural Fertility: Age Patterns, Levels, and Trends. In The Determinants of Fertility in Developing Countries. R.A. Bulatao and R.D. Lee, eds. Vol. I. New York: Academic Press, pp. 61-102. Knodel, John, Napaporn Havanon, and Anthony Pramualratana 1983 A Tale of Two Generations: A Qualitative Analysis of Fertility Transition in Thailand. Research Report No. 83-44, Population Studies Center, University of Michigan. Knodel, John, and Etienne van de Walle 1967 Breast Feeding, Fertility and Infant Mortality: An Analysis of Some Early German Data. Population Studies 21:109-31. 1979 Lessons from the Past: Policy Implications of Historical Fertility Studies. Popula· tion and Development Review 5:217-46. Kolff, G. H. van der 1936 The Historical Development of Labour Relationships in a Remote Corner of Java as they Apply to the Cultivation of Rice. New York: Institute of Pacific Relations. Koningsberger, V. J. 1948 De Eueopese Suikerrietcultuur en Suikerfabricatie. De Landbouw in de Indische Archipel, Vol. 2a, C. van Hall and C. van de Koppel, eds. Pp. 278-iOI. The Hquue: W. van Hoeve. Konnor, Melvin, and Carol Worthman 1980 Nursing Frequency, Gonadal Function, and Birth Spacing Amoq!Kuq Hunter· Gatherers. Science 207:788-91. Kuznets, Simon 1974 Rural-Urban Differences in Fertility: An International Comparison. Proceedinas of the American Philosophical Society 118:1-29. Laughlin, Charles Jr., and Ivan Brady, eds. 1978 Extinction and Survival in Human Populations. New York: Columbia University Press. Lee, P., et at. 1977 Endocrine and Metabolic Alterations with Food and Water Deprivation. American Journal of Clinical Nutrition 30:1953-62.

368 Lee, R. B. 1979 The !Kung San: Men, Women, and Work in a Foraging Society. Cambridge University Press. 1980 Lactation Ovulation, Infanticide, and Woman's Work: A Study of HunterGatherers Population Regulation. In "Biosocial Mechanisms of Population Regulation," M. N. Cohen, et. al., eds. New Haven: Yale University Press. Lee, Richard B., and Irven Devore, eds. 1968 Man the Hunter. Chicago: Aldine. Lee, Richard 1971 Decision Theory and Human Behavior. New York: Wiley. Leibenstein, H. 1974 An Interpretation of the Economic Theory of Fertility: Promising Path or Blind Alley? Journal of Economic Literature 12:457-79. 1981 Economic Decision Theory and Human Fertility Behavior. Population and Development Review 7:381-400. Leridon, H. 1977 Human Fertility: The Basic Components. Chicago: University of Chicago Press. Leridon, Henri, and Jane Menken, eds. 1979 Natural Fertility. Liege: Ordina. Lesthaeghe, Ron 1980 On the Social Control of Human Reproduction. Population and Development Review 6(4):527-48. Lesthaeghe, Ron, et a/. 1981 Child-Spacing and Fertility in Sub-Saharan Africa: An Overview of Issues. In ChildSpacing in Tropical Africa: Traditions and Change, Hilary J. Page and R. Lesthaeghe, eds. Pp. 3-23. New York: Academic Press. Levert, Ph 1934 Inheemse Arbeid in de Java Suikerindustrie. Wageningen: H. Veenman en Zonen. Levine, Robert A. 1976 Patterns of Personality of Africa. In Responses to Change. G. A. DeVos, ed. New York: Van Nostrand. 1980 Inftuence of Women's Schooling on Maternal Behavior in the Third World. Comparative Education Review 24:578-5105. Linares, 0. 1981 From Tidal Swamp to Inland Valley: On the Social Organization of Wet Rice Cultivation Among the Diola of Senegal. Africa 51:577-595. Lindert, Peter H. 1980 Child Costs and Economic Development. In Population and Economic Change in Developing Countries, Richard A. Easterlin, ed. Chicago: University of Chicago Press. Lobato, Rodolfo 1979 Quixin Quinal: La Colonizacion Tzeltal de Ia Selva Lacandon. Tesis de Licenciatura. Escuela Nacional de Antropologia, Mexico D.F. Lomnitz, Larissa 1978 Survival and Reciprocity: The Case of Urban Marginality in Mexico. In Extinction and Survival in Human Populations, Charles Laughlin, Jr., and Ivan Brady, eds. Pp. 181-297. New York: Columbia University Press. Longcope, C., J. Pratt, S. Schneider, and S. Feinberg 1978 Aromatization of Androgens by Muscle and Adipose Tissue in Vivo. Jounral of Clinical Endocrinology and Metabolism 46:146-52.

369 Longfield, Ada 1929 Anglo-Irish Trade in the Sixteenth Century. London: G. Routledge and Sons. Lough, Thomas 1896 England's Wealth, Ireland's Poverty. London: T. Fisher Unwin. Lovejoy, C. Owen 1981 The Origin of Man. Science 211:341-3SO. Luker, K. 197S Taking Chances: Abortion and the Decision Not to Contracept. Berkeley: University of California Press. Lunn, P. G., et at. 1980 Influence of Maternal Diet on Plasma-Prolactin Levels During Lactation. Lancet 1: 623-62S. MacDonald, M. M., and R. R. Rindfuss 1978 Relative Economic Status and Fertility: Evidence from a Cross-Section. In Research in Population Economics, Voi. 1, J. L. Simon, ed. Pp. 291-307. Greenwich, Connecticut: JAI Press. 1981 Earnings, Relative Income, and Family Formation: Demography 18:123-36. MacDonald, P., J. Grodin, and P. Siiteri 1969 The Utilization of Plasma Androstenedione for Estrone Production in Women. In Progress in Endrocrinology, C. Gual and F. Ebling, eds. Excerpta Medica, International Congress Series 184:770-76. MacDonald, P., R. Rombaut, and P. Siiteri 1967 Plasma Precursors of Estrogen I. Extend of Conversion of Plasma 4-androstenedione in Normal Males and Nonpregnant Normal, Castrate, and Adrenalectomized Females. Journal of Clinical Endocrinology and Metabolism 27:1103-11. MacGreil, Michael 1977 Prejudice and Tolerance in Ireland. Dublin: Research Section, College of Industrial Relations. Madawela, D. 1977 Human Resources in The Gambia: UNDP Mu1tidonor Mission to the Oambia River Basin. New York: UNDP. Maguire, Elizabeth, Ann A. Way, and Mohamed Ayad 1982 The Delivery and Use of Contraceptive Services in Rural Tunisia. International Family Planning Perspectives 8:96-101. Maguire, W. A. 1972 The Downshire Estates in Ireland, 1801-184S. Oxford: Clarendon Press. Malthus, Thomas 1963 Occasional Papers ofT. R. Malthus on Ireland, Population, and Political Economy from Contemporary Journals, Written Anonymously and Hitherto Uncollected. Bernard Semmel, ed. New York: Burt Franklin. Mamdani, Mahmood 1972 The Myth of Population Control. New York: Monthly Review Press. Mann, R. 197S The Oambia: Land Resource Degradation Survey. Banjul: Department of Agriculture, Yundum Experimental Station. March, J. G. 1978 Bounded Rationality, Ambiguity, and the Engineering of Choice. Bell Journal of Economics 9:S87-608.

370 Marshall, J., and C. Ritchie 1982 Where are the Nyae Nyae !Kung? Changes in the !Kung Economy 1958-1981 and the Introduction of Cattle. Manuscript. Masnick, G. S. 1979 The Demographic Impact of Breastfeeding. Human Biology 51:109-25. Masnick, GeorgeS., and Solomon H. Katz 1976 Adaptive Childbearing in a North Slope Eskimo Community. In Anthropological Studies of Human Fertility, Bernice Kaplan, ed. Detroit: Wayne State University Press (reprinted from Human Biology 48(1):1976). Mauldin, W. Parker 1980 Population Trends and Prospects. Science 209:148-57 May, B. 1979 The Indonesian Tragedy. London: Routledge. McAlpine, Phyllis J., and Nancy E. Simpson 1976 Fertility and Other Demographic Aspects of the Canadian Eskimo Communities of Igloolik and Hall Beach. In Anthropological Studies of Human Fertility, Bernice Kaplan, ed. Detroit: Wayne State University Press (reprinted from Human Biology 48(1): 1976). McBryde, F. W. 1970 Geograia Cultural e Historica del Suroeste de Guatemala. Guatemala: Seminario de Integration Social Guatemalteca. McCormick, James n.d. Professor of Community Health, Trinity College Medical School. Personal Communication. McDonald, P. 1980 An Historical Perspective to Population Growth in Indonesia. In Indonesia: Australian Perspectives, J. Fox, et. at., eds. Canberra: Research School of Pacific Studies, Australian National University. McEvoy, Frederick D. 1971 History, Tradition, and Kinship as Factors in Modern Sabo Labor Migrationn. Ann Arbor: University Microfilms. McGregor, I.. A., eta/. 1970 The Health of Young Children in a West African (Gambian) Village. Transactions of the Royal Society of Tropical Medicine and Hygiene 64:48-77. McKeown, Thomas 1976 The Modern Rise of Population. London: Arnold. McNeilly, A. 1979 Effects of Lactation on Fertility. British Medical Bulletin 35:15-154. Martin, L.G., J. Trussell, F.R. Salvail, and N.M. Shah 1983 Co-Variates of Child Mortality in the Philippines, Indonesia, and Pakistan: An Analysis Based on Hazard Models. Population Studies 37: 417-32. Meillasoux, C. 1973 Etate et Conditions des Esclaves Gubu (Mali) au XIXe, Siecle. Journal of African History 14:429-52. Melville, Margarita B., ed. 1980 Twice a Minority: Mexican American Women. St. Louis: C.V. Mosby. Menken, J., J. Trussell, and S. Watkins 1981 The Nutrition-Fertility Link: An Evaluation of the Evidence. Journal of Interdisciplinary History 11:425-41.

a

371 Messenger, John C. 1969 Inis Beag, Isle of Ireland. New York: Holt, Rinehart and Winston. Michael, R. T. 1973 Education and the Derived Demand for Children. In Economics of the Family: Marriage, Children, and Human Capital, T. W. Schultz, ed. Pp. 120-56. Chicago: University of Chicago Press, for the National Bureau of Economic Research. Milan, Federick A. 1970 The Demography of an Alaskan Eskimo Village. Artie Anthropology 7:26-43. Millard, Ann V. 1982 Child Mortality and Lactation Contraception in Mexico. Medical Anthropology 6:147-64. Mills, A. R. 1967 A Comparison of Urban and Rural Populations in the Lunsar Area: I. Lunsar, II. Village Populations. Sierra Leone Studies 20:173-90 and 21:12-51. Mincer, J. 1963 Market Prices, Opportunity Costs, and Income Effects. In Measurement in in Economics: Studies in Mathematical Economics and Econometrics in Memory of Yehuda Grunfeld, C. Christ and others, eds. Pp. 67-82. Stanford: Stanford University Press. Mitchel, John 1905 An Apology for the British Government in Ireland. Dublin: O'Donoghue and Co. Moghissi, K. S., and T. N. Evans, eds. 1977 Nutritional Impacts on Women. New York: Harper and Row. Mondot-Bernard, Jacqueline M. 1977 Relationships Between Fertility, Child Mortality and Nutrition in Africa. Paris: Development Center of the Organization for Economic Co-operation and Development. Morgan, Valerie 1976 A Case Study of Population Change over Two Centuries: Blaris, Lisburn 1661-1848. Irish Economic and Social History 3:5-16. Morton, A. L., and George Tate 1956 The British Labour Movement, 1770..1920: A History. London: Lawrence and Wishart. Murguia, Edward, and W. Parker Frisbie 1979 Trends in Mexican American Intermarriage: Recent Findings in Trends in Mexican American Intermarriage: Recent Findings in Perspective. Social Science Quarterly 58(3):374-89. Murray, Alice 1907 A History of the Commercial and Financial Relations between England and Ireland from the Period of the Restoration. London: P. S. King and Son. Nag, Moni 1962 Factors Affecting Human Fertility in Nonindustrial Societies: A Cross-cultural Study. Yale University Publieations·in Anthropology 66. 1972 Sex, Culture and Human Fertility in India and the United States. Current Anthropology 13:216-30. 1980 How Modernization Can Also Increase Fertility. Current Anthropology 21:571-87. Nag, Moni, and Neeraj Kak 1984 Demographic Transition in a Punjab Village. Population and Development Review 10: 661-78.

372 Nag, Moni, B. N. F. White, and R. C. Peet 1978 An Anthropological Approach to the Study of the Economic Value of Children in Java and Nepal. Current Anthropology 19:292-306. Namboodiri, N. K. 1975 Review of Economics of the Family. Demography 12:561-69. Nardi, Bonnie Anna 1981 Modes of Explanation in Anthropological Population Theory. American Anthropologist 83: 28-56. National Center for Health Statistics (NCHS) 1978 Facts of Life and Death. Hyatt$ville, Maryland: United States Department of Health, Education and Welfare. Nerlove, S. B. 1974 Work and Infant Feeding Practices: A Relationship with Demographic Implications. Ethnology 13:207-14. Netting, Robert M. 1965 Household Organization and Intensive Agriculture: the Kofyar Case. Africa 35(4):422-29. 1973 Fighting, Forest, and the Fly: Some Demographic Regulators Among the Kofyar. Journal of Anthropological Research 29:164-179. 1974 Agrarian Ecology. Annual Review of Anthropology 3:21-56. Newman, M. T. 1975 Nutritional Adaptation in Man. In Physiological Anthropology, Alfred Damon, ed. New York: Oxford University Press. Nimrod, A. and K. Ryan 1975 Aromatization of Androgens by Human Abdominal and Breast Fat Tissue. Journal of Clinical Endocrinology and Metabolism 40:367-72. Nisker, J., G. Hammond, and P. Siiteri 1980 More on Fatness and Reproduction. The New England Journal of Medicine 303:1124. Njoku, A. 0. 1979 The Economics of Mende Upland Rice Farming. In Essays on the Economic Anthropology of Liberia and Sierra Leone, V. R. Dorjahn and B. L. Isaac, eds. Pp. 103-120. Philadelphia: Institute for Liberian Studies. Nolan, William 1979 Fassadinin: Land, Settlement and Society in Southeast Ireland, 1600-1850. Dublin: Geography Publications. Notestein, Frank 1945 Population: The Long View. In Food for the World, Theodore Schultz, ed. Pp. 39-41. Chicago: University of Chicago. Oberg, Kalervo 1973 The Social Economy of the Tlingit Indians. Seattle: University of Washington Press. O'Connor, Kevin 1974 The Irish in Britain. Dublin: Tore Books. Odell, M. 1978 Family Formation, Productive· Systems and Demographic Change in Aguacatan, Guatemala. Unpublished doctoral dissertation, Temple University. 1982 The Domestic Context of Production and Reproduction in a Guatemalan Village. Human Ecology 10:47-69. O'Donovan, J. 1940 The Economic History of Livestock in Ireland. Dublin: Cork University Press.

373 Odunton, S., 0. Olu, 0. Ayeni, and 0. 0. Kale 1976 The Age of Menache in Nigerian Girls. Annals of Human Biology 3(3):269-74. O'Faolain, Sean 1954 Love Among the Irish. In The Vanishing Irish, John A. O'Brien, ed. London: W. H. Allen. O'Neill, Thomas 1956 The Organization and Administration of Relief, 1845-52. In The Great Famine: Studies in Irish History, 1845-52, R. Dudley Edwards and T. Desmond Williams, eds. Pp. 207-60. Dublin: Irish Committee of Historical Sciences. Oppenheimer, V. K. 1976 The Easterlin Hypothesis: Another Aspect of the Echo to Consider. Population and Development Review 2:433-57. O'Reilly, Kevin R. 1980 Population Dynamics and Family Planning in Dublin, Ireland. Unpublished doctoral dissertation, University of Connecticut. O'Rourke, John 1902 The History of the Great Irish Famine of 1847, with Notices of Earlier Famines. Dublin: James Duffy and Co. Ortiz de Montellane, Bernard 1978 Aztec Cannibalism: An Ecological Necessity? Science 200:611-17. Orwin, C., and E. Whetham 1964 History of British Agriculture, 1846-1914. Newton Abbot: David and Charles. Oshima, Harry 1983 The Industrial and Demographic Transitions in East Asia. Population and Development Review 9:583-608. O'Tuathaigh, Gear6id 1972 Ireland Before the Famine, 1798-1848.

Park, M. 1799 Travels in the Interior Districts of Africa Performed Under the Direction and Patronage of the African Association in the Years 1795, 1796, and 1797. London. Paul, A., E. Muller and R. Whitehead · 1979 The Quantitative Effects of Maternal Dietary Energy Intake on Pregnancy and

Lactation in Rural Gambian Women. Transactions of the Royal Society of Tropical Medicine and Hygiene 73:686-92. Perlman, M. 1975 Review of Economics of the Family. Demography 12:549-56. Peluso, N. 1979 Collecting Data on Women's Employment in Rural Java. Studies in Family Planning 10:374-78. Penders, C. L. M. 1977 Indonesia. Brisbane: University of Queensland Press. Peper, B. 1970 Population Growth in the Nineteenth Century: A New Interpretation. Population Studies 24:71-84. Perel, E., and D. Kinninger 1979 The Interconversion and Aromatization of Androgens by Huinan Adipose Tissue. Journal of Steroid Biochemistry 10:623-27. Peterson, Nicolas, ed. 1976 Tribes and Boundaries in Australia. Canberra: Australian Institute of Aboriginal Studies, Social Anthropology.

374 Poel, H. A. van der 1865 Nota over de Rijstkultuur op Java. Tijdschrift voor Nijverheid en Landbouw in Nederlandsch-Indie 11 :97-118. Polgar, Steven 1964 Evolution and the Ills of Mankind. In Horizons of Anthropology, Sol Tax, ed. Chicago: Aldine. 1972 Population History and Population Policies from an Anthropological Perspective. Current Anthropology 13:203-11. 1975 Cultural Development, Population, and the Family. In The Population Debate. Papers of the World Conference, Bucharest, 1974. Vol. II. United Nations. Porteres, Alejandro 1976 On the Sociology of National Development. American Journal of Sociology 85:55-84. Prest, John 1972 Lord John Russell. London: Macmillan. Preston, Samuel H. 1975 Health Programs and Population Growth. Population and Development Review 1:89-200. Prinsen-Geerligs, P. H. 1914 The Java Sugar Cane Industry. Den Hagg: Van Dopp. Pullum, T. W. 1980 Separating Age, Period and Cohort Effects in White U.S. Fertility, 1920-1970. Social Science Research 9P:225-44. Purves, George (Simon Gray) 1818 Gray versus Malthus: The Principles of Population and Production Investigated. London: Longman, Hurst, Rees, Orme and Brown. Quinn, C.

1972 Mandingo Kingdoms of Senegambia. Evanston: Northwestern University Press. Raffles, T. S. 1817 The History of Java. London: John Murray. Rapoport, A., and T. S. Wallsten 1972 Individual Decision Behavior. In Annual Review of Psychology, Vol. 23, P. H. Mussen and M. R. Rosenzweig, eds. Pp. 131-76. Palo Alto: Annual Review, Inc. Rattray, Robert Sutherland 1932 The Tribes of the Ashanti Hinterland. 2 Volumes. London: Oxford University Press. Recinos, A. 1954 Monografia de Departamento del Huehuetenango, Guatemala. Guatemala: Editorial del Ministerio de Educacion Publica. Redford, Andrew 1926 Labour Migration in England, 1800-50. Manchester: Manchester University Press. Rens, M. C. 1980 Breastfeeding and Female Employment in Rural Java. Unpublished dissertation, University of California at Davis. Retei-Laurentin, Anne, and Daniel Benoit 1976 Infant Mortality and Birth Intervals. Population Studies 30:279-293. Retherford, R.D., and J.A. Palmore 1983 Diffusion Processes Affecting Fertility Regulation. In The Determinants of Fertility in Developing Countries. R.A. Bulatao and R.D. Lee, eds. Vol. 2. Pp. 295-349. New York: Academic Press.

375

Romaniuk, A. 1981 Increase in Natural Fertility During the Early Stages of Modernization: Canadian Indians Case Study. Demography 18:157-72. Rose Innes, R. 1977 A Manual of Ghana Grasses. Surrey, England: Land Resource Division, Ministry of Overseas Development. Rose, R. S. 1976 An Outline of Fertility Control, Focusing on the Element of Abortion in the Republic of Ireland to 1976. A Thesis presented to the Institute of Sociology, The University of Stockholm, Stockholm, Sweden. Ross, Eric i.p. The Riddle of the Scottish Pig. BioScience. Rowland, M. G. M., eta/. 1980 Parity and Lactation Performance in Gambian Mothers. Lancet 11:1365. 1981 Seasonality and the Growth of Infants in a Gambian Village. In Seasonal Dimensions to Rural Poverty, R. Chambers, eta/., eds. Pp. 164-75. Totawa, New Jersey: t\llanheld, Osmun and Company. Russell, Edward John 1959 Agriculture-Great Britain. Encyclopaedia Britan-nice, Vol. 1:357-62. Ryder, N. B. 1973 Comment. In Economics of the Family: Marriage, Children, and Human Capital, T. W. Schultz, ed. Pp. 76-80. Chicago: University of Chicago Press, for the National Bureau of Economic Research. Sabagh, Georges, and David Lopez 1980 Religiosity and Fertility: The Case of Chicanas. Social Forces 59(2):431-37. Salaman, Redcliffe 1943 The Influence of the Potato on the Course of Irish History. Dublin: The Richview Press. Salber, E., M. Feinleib, and B. MacMahon 1965 The Duration of Postpartum Amenorrhea. Journal of Epidemilogy 82:347-58. Sanderson, W. C. 1980 Comment. In Population and Economic Change in Developing Countries, R. A. Easterlin, ed. Pp. 140-44. Chicago: University of Chicago Press. Santow, G. 1978 A Simulation Approach to the Study of Human Fertility. The Hague: Nijhoff. Saucier, J. 1972 Correlation of the Long Post-partum Taboo: A Cross-Cultural Study. Current Anthropology 13:238-42. Schumann, Debra 1982 Fertility and Economic Strategy in a Mexican Ejido. Ann Arbor: University Microfilms. Schutjer, W.A., and C.S. Stokes 1982 The Human Fertility Implications of Food and Agricultural Policies in Less Developed Countries (Bull. 835). University Park, Pennsylvania: College of Agriculture, Pennsylvania State University. Schwab, George 1947 Tribes of the Liberian Hinterland. Papers of the Peabody Museum of American Archeology and Ethnology, Harvard University. Vol. 31. Cambridge: Peabody Museum.

376 Schneider, Jane, and Peter Schneider 1984a Demographic Transition in a Sicilian Rural Town. Journal of Family History Fall:245-72. 1984b The Demographic Transition of Landpoor Peasants in a Sicilian Rural Town: A Study of Class and Family. Paper presented to the 85th Annual Meeting of the American Anthropological Association, Denver. Schultz, Theodore W. 1977 On Economic History in Extending Economics. Economic Development and Cultural Change 25 (Supplement):245-53. Scrimshaw, Susan C. M. 1978a Stages in Women's Lives and Reproductive Decision-Making in Latin America. Medical Anthropology 2:41-58. 1978b Infant Mortality and Behavior in the Regulation of Family Size. Population and Development Review 4:383-404. Selosoemardjan 1962 Social Changes in Jogjakarta. Ithaca: Cornell University Press. Shelton, Walter 1973 English Hunger and Industrial Disorder. London: Macmillan. Short, R. 1976 Lactation-the Central Control of Reproduction. In Breast-feeding and the Mother. Excerpta Medica- North Holland, Ciba Foundation Symposium 45:73-86. Shorter, Edward 1975 The Making of the Modem Family. New York: Basic Books. Shuttleworth-Kay, James 1969 [orig. 1832] The Moral and Physical Condition of the Working Classes Employed in the Cotton Manufacture in Manchester. Manchester: E. J. Morten. Sierra Leone, Government of 1953 Atlas of Sierra Leone. Freetown: Survey and Lands Department. 1965 1963 Population Census of Sierra Leone. Freetown: Central Statistics Office. Sievers, A. 1974 The Mystical World of Indonesia: Culture and Economic Development in Conflict. Baltimore: Johns Hopkins University. Simon, H. A, 1980 The Behavioral and Social Sciences. Science 209:72-78. Simon, Julian L. 1974 The Effects of Income on Fertility. Chapel Hill: Carolina Population Center. 1977 The Economics of Population Growth. Princeton: Princeton University Press. Singarimbun, M., and C. Manning 1974 Fertility and Family Plarming in Mojolama. Yogyakarta: Lembaga Kependudukan. Slovic, P., B. Fischhoff, and S. Lichtenstein 1977 Behavioral Decision Theory. In Annual Review of Psychology, Vol. 28., M. R. Rosenzweig and L. W. Porter, eds. Pp. 1-39. Palo Alto: Annual Reviews, Inc. Smart, William 1964 [orig. 1910] Economic Annals of the Nineteenth Century, 1801-1820, Vol. 1. New York: Augustus Kelley. Smetana, J. G., and N. E. Adler 1979 Decision-making Regarding Abortion: A Value X Expectancy Analysis. Journal of Population 2:338-57. Smith, Antony H. 1981 Sociocultural Factors in Child Mortality Among the Ternne of Magburaka. Unpublished doctoral dissertation, University of Oregon.

377

Smith, Carol 1977 How Marketing Systems Affect Economic Opportunity in Agrarian Societies. In Peasant Livelihood, R. Halperin and J. Dow, eds. Pp. 117-146. New York: St. Martins Press. Smith, D.P. 1981 A Reconsideration of Easterlin Cycles. Population Studies 35:247-64. Smith, Jack C., eta/. 1982 Trends in the Incidence of Breastfeeding for Hispanics of Mexican Origin and Anglos on the US-Mexican Border. Public Health Briefs, AJPH, January, Vol. 72, No.I. Smith, M. B. 1978 Decision Making and Psychological Antecedents: A Discussion. In The First Child and Family Formation, W. B. Miller and L. F. Newman, eds. Pp. 266-72. Chapel Hill: Carolina Population Center. Soest, G. H. van 1869 Geschiedenis van het Kultuurstelsel. Rotterdam: Nijgh. Solow, Barbara 1971 The Land Question and the Irish Economy, 1870-1903. Cambridge: Harvard University Press. Spaulding, E., et a/. 1977 A Study of Severely Malnourished Children in the Gambia. Journal of Tropical Pediatric Environmental Child Health 23:215-19. Spencer, Robert F. 1979 The North Alaskan Eskimo. Smithsonian Institution, Bureau of American Ethnology Bulletin 171. Washington, D.C.: U.S. Government Printing Office. Stamm, Liesa, and Amy Ong Tsui n.d. Life Course Effects in Marriage Patterns among Women in a Tunisian Town. Unpublished Manuscript. Stamp, Dudley, and Stanley Beaver 1971 The British Isles: A Geographic and Economic Survey. London: Longman. Stanik, S., et al. 1981 The Effect of Weight Loss on Reproductive Hormones in Obese Men. Journal of Clinical Endocrinology and Metabolism 53:828-32. Steele, E. D. 1974 Irish Land and British Politics: Tenant-Right and Nationality, 1865-1870. London: Cambridge University Press. Stein, Z., M. Susser, G. Saenger, and F. Marolla 1975 Famine and Human Development. New York: Oxford. Stephen, Leslie, and Sidney Lee, eds. 1921-22 The Dictionary of National Biography. London: Oxford University Press. Steward, Julian 1938 Basin-Plateau Aboriginal Sociopolitical Groups. Smithsonian Institute Bureau of American Ethnology Bulletin. Stokes, Whitley 1821 Observations on the Population and Resources of Ireland. Dublin: Joshua Porter. Stycos, J. Mayone 1963 Problems of Fertility Control in Under-Developed Areas. Marriage and Family Living 25:5-13. Sussman, R. W. 1972 Child Transport, Family Size, and Increase in Human Population During the Neolithic. Current Anthropology 13:258-9.

378 Sweezy, A. 1971 The Economic Explanation of Fertility Changes in the United States. Population Studies 25:255-67. Swift, Jonathan 1739-40 Elegy on the Much Lamented Death of Those Excellent Patriots and Lovers of their Country the Family of the Potatoes in the Kingdom of Ireland. Dublin: George Faulkner. Syme, J. G. G. 1932 The Kusasis. A Short History. Manuscript (in my files). Tabah, Leon 1980 World Population Trends, a Stocktaking. Population and Development Review 6:355-90. Tax, Sol 1953 Penny Capitalism: A Guatemalan Indian Economy. Washington: The Smithsonian Institution. Taylor, Carl E., Jeanne S. Newman, and Narindar U. Kelly 1976 The Child Survival Hypothesis. Population Studies 30:263-77. Texa~ Industrial Commission 1977 General Community Prome of Rio Grande City. Austin, Texas. Texas Industrial Commission 1981 General Community Prome of Rio Grande City. Austin, Texas. Thadani, Veena N. 1978 The Logic of Sentiment: The Family and Social Change. Population and Development Review 4:457-99. The Gambia 1974 Population Census: 1973 (Vol. I). Banjul: Central Statistics Division. The London Times 1880 The Great Irish Famine of 1845-46. London: The Times. Thernstrom, Stephen, ed. 1980 Harvard Encyclopedia of American Ethnic Groups. Cambridge: Harvard Univ. Press. Thomas, Northcote W. 1916 Anthropological Report on Sierra Leone. London: Harrison and Sons. Thomas, Roger A. 1973 Forced Labor in British West Africa: The Case of the Northern Territories of the Gold Coast 1906-1927. Journal of African History 14(1):79-103. Thompson, L., and G. B. Spanier 1978 Influence of Parents, Peers, and Partners on the Contraceptive Use of College Men and Women. Journal of Marriage and the Family 40:481-92. Thompson, Robert 1802 Statistical Survey of the County of Meath, with Observations on the Means of Improvement. Dublin: Graisberry and Campbell. Thornton, A. 1978 The Relationship Between Fertility and Income, Relative Income, and Subjective Well-Being. In Research in Population Economics, Vol. I, J. L. Simon, ed. Pp. 261-290. Greenwich, Connecticut: JAI Press. 1979 Fertility and Income, Consumption Aspirations and Child Quality Standards. Demography 16:157-75. Tietze, Christopher 1976 The Effect of Breast Feeding on the Rate of Conception. Proceedings of the International Population Conference, New York 2:129-36.

379 1979 Legal Abortion in the World of Today. In Pregnancy Termination, Gerald Zatuchni, John Sciarra and J. Speidel, eds. Pp. 406-16. New York: Harper and Row. Tiffen, M. 1975 The Enterprising Peasant. London: HMSO. Tighe, William 1802 Statistical Observations Relative to the County of Kilkenny, Made in the Years 1800 and 1801. Dublin: Graisberry and Campbell. Townes, B. D., F. L. Campbell, L. R. Beach, and D. C. Martin 1976 Birth-Planning Values and Decisions. In Population Psychology: Research and Educational Issues, S. H. Newman and V. D. Thompson, eds. Pp.ll3-30. Washington, D.C.: U.S. Government Printing Office. 1977 Birth Planning Values and Decisions: The Prediction of Fertility. Journal of Applied Social Psychology 1:73-88. Townes, B. D., L. R. Beach, F. L. Campbell, and R. L. Wood 1980 Family Building: A Social Psychological Study of Fertility Decisions. Population and Environment 3:210-20. Trevelyan, Charles 1848 The Irish Crisis. London: Longman, Brown, Green and Longmans. Trevelyan, George Macaulay 1942 A Shortened History of England. Harmondsworth: Penguin. Trussell, James 1979 Natural Fertility: Measurement and Use in Fertility Models. In Natural Fertility, Henri Leridon and Jane Menken, eds. Pp. 29-64. Liege: Ordina. Truswell, A. S., and J. D. L. Hansen 1976 Medical Research Among the !Kung. In Kalahari Hunter-Gatherers, R. B. Leeand I. DeVore, eds. Pp. 167-194. Cambridge: Harvard University Press. Tsui, A. 0., et at. 1979 Sierra Leone. Illustrative Functional Projections, 1975-2000. Chicago: Community and Family Study Center, The University of Chicago. Tucker, G.S.L. 1970 Irish Fertility Ratios Before the Famine. Economic History Review 23:267-84. Tuke, James 1880 Irish Distress and Its Remedies; The Land Question; A Visit to Donegal and Connaught in the Spring of 1880. London: W. Ridgway. Turchi, B. A. 1975a Microeconomic Theories of Fertility: A Critique. Social Forces 54:107-25. 1975b The Demand for Children: The Economics of Fertility in the United States. Cambridge: Ballinger. Turke, Paul W. 1984 On What's Not Wrong with a Darwinian Theory of Culture. American Anthropologist 86:663-668. Turnbull, Colin 1972 The Mountain People. New York: Simon and Schuster. United Nations 1982 World Population Trends and Policies, 1981 Monitoring Report. Volume 1: Population Trends. Department of International Economic and Social Affairs. Population Studies No. 79. New York: United Nations. (Tables 5 and 29). UNDP Multidisciplinary Multidonor Mission 1980 Development of the Gambia River Basin: Final Mission Report (Chapters I-IV). New York: UNDP.

380 United States Census Bureau 1970 U.S. Census of Population. "General Social and Economic Characteristics." P.C. (1) - C 45, Texas. U.S. Department of Commerce (Bureau of the Census, Population Division) 1977 Ghana. Country Demographic Profiles. ISP-DP-5. Washington, D.C.: U.S. Department of Commerce, Bureau of the Census. Ussher, Arland 1954 The Boundary Between the Sexes. In The Vanishing Irish. John A. O'Brien, ed. London: W. H. Allen. Uzzell, J. Douglas 1980 Mixed Strategies and the Informal Sector: Three Faces of Reserve Labor. Human Organization 39: 40-49 .. Van Niel, R. 1969 Sugar Cultivation in Pasuruan, Java. Journal of Oriental Studies 7:261-76. 1972 Measurement of Change under the Cultivation System in Java 1837-185 1. Indonesia 14:89-109. Vaughan, W. E., and A. J. Fitzpatrick 1978 Irish Historical Statistics: Population, 1821-1971. Dublin: Royal Irish Academy. Veer, K. van der 1948 Rijst. In De Landbouw in de Indische Archipel, Vol. 2a., C. van Hall and C. van de Koppel, eds. Pp. 7-109. The Hague: W. van Hoeve. Velez, I. Carlos G. 1980 The Nonconsenting Sterilization of Mexican Women in Los Angeles: Issues of Psychocultural Rupture and Legal Redress in Paternalistic Behavioral Environments. In Twice a Minority: Mexican American Women, Margarita B. Melville, ed. St. Louis: C.V. Mosby. Verdcllen J. 1973 The Ecological Basis of Agriculture and Production Patterns in the Upper River Division of the Gambia. Berlin: World Friendship (Basse, The Gambia: Community Services). Vigersky, R. A. Anderson, R. Thomson, and D. Loriaux 1977 Hypothalamic Dysfunction in Secondary Amenorrhea Associated with Simple Weight Loss. New England Journal of Medicine 297:1141-45. Vincent, P. 1950 La Sterilite Physiologique de Populations. Population 5:45:64. Vink, G. J. 1941 De Grondslagen van het Indonesische Landbouw Bedrijf. Wageningen: H. Veenman en Zonen. Vitalis, L. 1841 [1977] De invoering, werking en gebreken van het stelsel van Kultures op Java. Indonesia: C. L. M. Penders, translation. Brisbane: University of Queensland Press. Vries, E. de 1931 Landbouw en Welvaart in het Regentschap Pasoeroean: Bijdrage tot de Kennis van de Sociale Economie van Java, 2 Vols. Wageningen: H. Veenman en Zonen. Wachter, M. L. 1975 A Time Series Fertility Equation: The Potential for a Baby Boom in the 1980s. International Economic Review 16:609-24. Wakefield, Edward 1812 An Account of Ireland, Statistical and Political. 2 Vol. London: Longman, Hurst, Rees, Orme and Brown.

381 Wallsten, T. S. (ed.) 1980 Cognitive Processes in Choice and Decision Behavior. Hillsdale, J.J.: Lawrence Erlbaum Associates. Walsh, Brendan M. 1974 Ireland. In Population Policy in Developed Countries, B. Buelson, ed. Pp. 9-41. New York: McGraw-Hill. Ward, J. T. 1970 The Factory System. Vol I: Birth and Growth. Netton Abbot: David and Charles. Wei!, P.M. 1970 Introduction of the Oxplow in Central Gambia. In African Food Production Systems: Cases and Theory, P. F. M. McDoughlin, ed. Pp. 231-63. Baltimore: The Johns Hopkins Press. 1973 Wet Rice, Women and Adaptation in the Gambia. Rural Africana, Winter:21-9. 1976 The Staff of Life: Food and Female Fertility in a West African Society. Africa 46:182-95. 1978 Social Analysis of Alternative Development Strategies for the Gambia River Basin: 1976-2000. Washington: Robert R. Natha, Associates (for Sahel Development Program, USAID/Washington). 1980 Mandinka Adaptation to Colonial Rule in the Gambia. Cultures et Developpement 12:295-318. 1981 Human Resources and Socio-Economic Constraints: The Lower Casamance Master Plan Project. Chicago: Harza Engineering Company (for USAID and the Government of Senegal). 1984 Slavery, Groundnuts, and European Capitalism in the Wuli Kingdom of Senegambia: 1820-1930. Research in Economic Anthropology, 6. Wells, J. C. 1980 Review of Agricultural Change in Tropical Africa, K. R. M. Anthony eta/. African Economic History 9:121-125. Wertheim, W. F. 1950 Effects of Western Civilization on Indonesian Society. New York: Institute of Pacific Relations. 1959 Indonesian Society in Transition. The Hague: W. van Hoeve. Westhoff, Charles 1978 Marriage and Fertility in the Developed Countries. Scientific American 239:51-7. Westoff, Charles F., et a/. 1961 Family Growth in Metropolitan America. Princeton: Princeton University Press. Whelan, Brendan n.d. Ransam: A National Random Sample Design for Ireland. White, B. 1973 Demand for Labor and Population Growth in Colonial Java. Human Ecology 1:217-36. 1974 Reply to Geertz and van de Walle. Human Ecology 2:63-5. 1976 Population, Employment and Involution in Rural Java. Development and Change 7:267-90. White, D. R., and M. L. Burton 1981 Sexual Division of Labor in African Agriculture: A Network Auto Correlation Analysis. American Anthropologist 83:824-49. Whitehead, R., eta/, 1978 Factors Influencing Lactation Performance in Rural Gambian Mothers. Lancet 2:178-81.

382 Whiteford, Linda M. 1979 Family Relations in Seco County: A Case Study of Social Change. Unpublished doctoral dissertation, University of Wisconsin, Milwaukee. 1980 Mexican American Women as Innovators. In Twice a Minority. Mexican American Women, Margarita Melville, ed. St. Louis: Mosby Press. 1982 Migrants No Longer. De Colores Journal of Chicano Expression and Thought, Vol. 6, Nos. 1 and 2. Widjojo, N. 1970 Population Trends in Indonesia. Ithaca: Cornell University Press. Wilk, Richard n.d. Labour Management and Household Organization Among the Kekchi Maya. Manuscript available frpm the author, Department of Anthropology, University of Arizona. Willis, R. J. 1973 Economic Theory of Fertility Behavior. In Economics of the Family: Marriage, Children, and Human Capital, T. W. Schultz, ed. Pp. 25-75. Chicago: University of Chicago Press, for the National Bureau of Economic Research. 1982 The Direction of Intergenerational Transfers and Demographic Transition: The Caldwell Hypothesis Reexamined. Population and Development Review 8 (Supplement):207-34. Wilmsen, E. 1978 Seasonal Effects of Dietary Intake on Kalahari San. Federation Proceedings 37:65-72. 1981 Exchange, Interaction, and Settlement in Northwestern Botswana: Past and Present Perspectives. In Proceedings of the Symposium on Settlement in Botswana, Renee Hitchcock and M. Smith, eds. Pp. 98-109. Heinemann. 1982 Studies in Diet, Nutrition, and Fertility among a Group of Kalahari B\IShmen in Botswana. Social Science Information 22:95-125. n.d. Variation in Circulating Lipids and Reproductive Hormones Associated with Diet in Kalahari Bushmen. In preparation. Wilson-Davis, Keith 1975 Some Results of an Irish Family Planning Survey. Journal of Biosocial Science 7:435-44. Winikoff, B. 1978 Nutrition, Population and Health. Science 200:895-902. Woodburn, James 1968 An Introduction to Hadza Ecology. In Man the Hunter, R. B. Lee and I. DeVore, eds. Chicago: Aldine. Woodham-Smith, Cecil 1962 The Great Hunger: Ireland, 1845-49. New York: E. P. Dutton. Wolfers, David, and Susan Scrimshaw 1975 Child Survival and Intervals Between Pregnancies in Guayaquil, Ecuador. Population Studies 29(3):479-496. World Bank 1982 World Development Report 1982. New York: Oxford University Press. 1983 World Development Report 1983. New York: Oxford University Press. Wynne-Edwards, V. C. 1962 Animal Dispersion in Relation to Social Behavior. Edinburgh: Oliver and Boyd. Wray, J. 1978 Maternal Nutrition, Breast-feeding, and Infant Survival. In Nutrition and Human Reproduction, W. Moseley, ed. Pp. 197-229.

INDEX Abortion incuded 9, 94, 103-104, 142-143, 156-158, 170-171, 251, 269, 273, 295, 309, 332 natural, miscarriage, 94, 104, 116, 118, 135, 332 See also Fertility, proximate determinants of Abstinence post-partum, 6, 72, 91, 95, 100, 103, 105, 107, 189, 249, 251-253, 258, 260-262, 265, 273, 283-285, 287, 295, 305, 312 See also Fertility, proximate determinants of Africa, 1, 2, 18, 101, 319, 346 Angola, 66 Botswana, 66 Cameroon, 346 Gabon, 336 Gambia, 21, 74, 294-320 Ghana, 21, 263-293, 338 Ivory Coast, 319, 346 Kenya, 336 Liberia, 18, 23, 90-111, 338 Mali, 319 Namibia, 66 Nigeria, 319 Senegambia, 319 Sierra Leone, 2, 319, 321-349 West Africa 1, 2, 156, 321, 337 Age in census enumerations, 91, 323 dating, 91-93, 113, 268, 323-326 Age specific fertility ratios, 66-68, 115, 118, 121, 229 Age pattern of childbearing, 4-5, 9, 115, 119, 123-124, 132, 156, 242-243, 245, 247 Agriculture, 92, 127, 191, 238, 240, 263, 277 commercial, 92-93, 96-100, 128-130, 146, 148-151, 157, 209, 294, 298, 313 Cultivation System, 253-255, 259, 261-262 intensification, 129-130, 140, 253-255, 278-280, 291, 294, 298-304, 313, 316-318, 320 strategies, 127, 130, 300, 304-305, 313, 316 swidden, upland, 92, 130, 277-280, 298-299 Amenorrhea, 107, 260 lactation-induced, 1, 8, 72-77, 107, 273, 283-284, 312 dietary, 1, 8, 73-77 interaction between lactation and diet 1, 8, 72-83 See also Breastfeeding, Fertility, proximate determinants of, Lactation, Nutrition Artisans, craftsmen, 19, 131, 299 Bangladesh, 74, 260, 291 Bassa, 91, 101

383

384 Belize, 15 Birth control (See Abortion, Contraception, Fertility, proximate determinants of) Birth, 103, 296 illegitimate, 63 intervals, 66, 69, 121, 132, 169-171, 265, 273, 275, 283, 287, 309 seasonality, 113-114, 116-117, 308-309 Bottle feeding, 112, 116, 123, 132 Breastfeeding, 72-73, 283-284 as contraceptive method, 4-5, 7, 72-77, 103, 121, 156,241,251-253,258,260-262, 283-284, 295, 312, 332 changes in, 107, 253, 295, 305, 312 duration of, 105-107, 132, 253 See also Fertility, proximate determinants of, Lactation Catholicism/Catholic, 11, 132, 147, 157-158, 198, 209, 211, 214-215, 221, 223-227, 230-233, 235 Child-care quality of, 108-110, 189-190, 310 Children aspirations for, 15, 26, 246 costs of, 15, 17, 102-104, 127, 130-131, 154, 157, 244-247, 256, 291, 339-346 desire/demand for, 1, 5-7, 91, 107-110, 132, 144, 173, 280, 291 labor, 102, 129-131, 149-151, 211, 213, 244 prestige value, 1, 162-163, 173-174 as productive assets, 14-16, 18-22,26, 102, 127, 129-131, 150-151, 157, 163, 173174, 190, 244, 247-248, 256, 277, 279-280, 291, 304, 320, 339-346 supply, 5-7, 21, 265, 277, 285-287 China, 16, 192, 318-319 Coast Salish, 190 Coital frequency, 90, 108, 141-142, 151, 309, 312, 332, 337 Colonial, colony, 197-203, 207, 210, 214-216, 248-256, 258-259, 261-262, 276 Competition, 176-195, 207, 214 and reproductive behavior, 189-195 Contraception 4-6, 8-9, 22-23, 43-55, 96, 100, 103, 121, 126, 132, 142-143, 158, 171, 222, 226-230, 235-237, 239, 242, 245, 269, 295, 309, 332, 336 Cooperation, 177-178, 180-181, 183, 186, 188, 191 Cottiers, 199-201, 207, 210, 213 Cree, 180 Culture, 10 and biology, 25-27 definition of, It and fertility, 10-12 and economic/sociallpolitical/religious/psycho1ogical factors, 11 and decision-making, 11-12 as independent variable, 10-12, 14, 24, 101, 157-158, 176, 218, 221, 223, 230-234, 237, 240-242, 244-248, 337-346 as dependent variable, 11-12,25, 102-108, 113, 115-116, 119, 121, 176, 191,218,

385 230-234, 240, 265, 291, 294 Cultural materialism, 25, 191 Dagomba, 275 Decision-making models of, 2, 11-12, 24, 30-57, 287, 291 and fertility, 11-12, 30-57, 108-109, 132, 144, 156-159, 230, 234, 236, 238, 242, 245-247, 257-258, 320, 345 See also Rationality Demand for children (See Children, desire/demand for) Demographic transition modernization theory of, 3-7 criticisms of, 7-10, 25 wealth flows theory of, 3, 12-13, 127, 154, 339 criticisms of, 14-15, 191-192, 223, 225 Diet, 66-67, 72-73, 79-82, 113, 115-116, 118, 199, 201, 213, 250, 279, 334 See also Food, Nutrition Divorce, 64, 225, 246, 253, 280, 295 Dutch, 249-251, 253-256, 258, 262 Economic development, 112-113, 115-116, 128-131, 145-146, 161-162, 197-201, 239241, 297-304, 313, 321-323, 346-348 Economic niche, strategy, 1-2, 127, 129-131, 144-146, 148-151, 154, 156-158, 209, 238, 242-248, 300, 347 Economic structure, 147-148, 240, 248 Economics of fertility, 5-6, 125-127, 287, 291 Education, 94, 96, 104, 130, 147, 166, 168-169, 223, 234, 241-242, 244, 246, 279, 296, 327 332-333, 338, 344 and fertility, 6-7, 17-20, 100, 104, 223-225, 338-340 and mortality, 21, 23 Enrlgration, 128,169,190,202,215,217,222,228,258,263,265,276-277,279,303304, 306, 309, 319 Employment opportunities and fertility, 1-3, 5, 15-17, 19, 131, 225-226, 240-241, 248 England, 197-217 Eskimo, 1, 23, 112-124 Europe, 101, 259, 319 Western Europe, 7, 201, 221 Evolutionary theory, 25-27 t

Family planning programs/services, 90, 110, 121, 157-159, 162, 221, 226-228, 230, 234, 236, 269 and fertility 22-23 Family size 3 Fecundability 1, 59, 72-81, 139, 142, 309 Fecundity 1, 59, 81, 139, 169, 273

386 Fertile Crescent, 192 Fertility control technologies, 4-6, 9, 123-124, 159, 170-171 decline, 197, 213, 215, 222-223, 312, 319 development and, 5-6 historical, 3 increase, 112, 118, 123, 197, 249, 263, 268-269, 273, 306, 312-313, 320 indicators of 6-10, 13 proximate determinants of, 4-6, 8, 59-89, 294, 320 Bongaarts' model of, 8-9, 59-69, 70-81, 83-90, 119, 121, 265, 169, 273, 280, 283, 330-332 natural, 4-5, 9, 59-89, 90-91, 94, 109-111, 119-121, 127, 131-134, 139, 151, 309 models of, 99, 101, 134-140, 151, 154 rural, 2, 323, 327, 329-330 target, S-6, 9, 238, 309 . transition, 1-27, 112, 121, 123-124, 221, 225 urban, 2, 18, 323, 327, 329-330 urban-rural differences, 2, 323, 327, 329-330 Food, 104, 204-206, 261, 298-299, 313 prices, 207, 301-303, 316, 342, 347 supply, 112-113, 130, 200, 203, 296-297, 304, 334 See also Diet, Potato, Resources Foragers (See Hunters and gatherers) Fularti, 317 Gi:eat Basin, 180 Guatemala, 1, 74, 83-84, 89, 125-143 Haida, 190 Health, 285, 296 care, 139-141, 154, 157-158, 189, 226, 268, 284 services, 92, 110-111, 128, 141-142, 239, 285, 297 Hunters and gatherers, 59-89, 112-124, 177-185, 189-192 Immigration, 146-147, 186, 229, 255-256, 332 Income, 6, 93, 96, 100, 240-241, 296, 313 Industrialization, 3-4, 177, 193-194, 207-213 Industrial revolution, 3-4, 186-188 Infanticide, 21-22, 154, 189-190 Infertility, 68, 104, 110, 132, 312 See also Sterility Intermediate fertility variables (See Fertility, proximate determinants of) Ireland, 2, 7, 14, 196-236 Java, 156, 249-262, 291 Kinship, 101, 189-190, 194, 275, 277, 280, 304-306, 333 Korea, 88

387 Kpelle, 92, 101 Kusasi, 263-293 Labor age of entry, 102, 149, 163, 277, 304 division of, 102, 119, 149, 244-247, 257-260, 277, 294, 297-305, 316-320 landless, 19-20, 131, 135, 206, 214, 216 niigratory wage, 129-130, 142, 154, 240-242, 244-245, 303-304, 312-313, 319 productivity, 207 shortage, 253-255, 279, 300, 305, 319 tenants, 135, 199-200, 209, 211, 216 wage, 148, 150, 157 Lactation, 21, 66-70, 95, 100, 105, 151, 156, 171, 273, 297, 309 customs, 21, 105, 132, 136 change in, 21, 107 See also Amenorrhea Land

inheritance, 131 rents, 198, 202, 207, 216 shortage, 210, 278 tenure, 199, 215 Landed gentry, landowners, 19, 130, 142-143, 198-199, 203, 207, 209, 211, 218, 223 Latin America, 11, 101, 188 Lorna, 91, 101

Macro, 1, 2, 24-27 Malthus/Malthusian, 2, 196-197, 203-207, 213-214, 217-218, 222, 250-251, 253, 338, 346

Mamprussi, 275-276 Mandinka, 294-295, 298, 303, 305-306, 309, 312, 316-317, 320 Mano, 92, 101 Manupur,23 Marriage, 93-94, 107, 131-132, 163-169, 215, 280, 283, 295, 318 age at, 64, 91, 94, 100, 151, 165-166, 169, 211, 213, 253, 262, 273, 335 change in, 168-170, 211, 213, 222-223, 229, 273, 276, 283, 312 European marriage pattern, 223 and fertility, 9, 59-64, 66, 69-72, 107-143, 229, 283 Material well-being/welfare, 3, 14-17, 19-23 relative prosperity, 16-17 Mass media, 6 Maya, Tzeltal, 144-158, 181 Mesoamerica, 192 Mennonite, 188 Mexico, 1, 144-158 Micro 1-2, 24-27 Microeconomic theory, 31-35 Middle East, 167-170

388 Mexican Americans, 2, 18, 237-248 Modernization 3, 6-8, 17, 25 Morality, 13-15 Mortality, 295-296, 306, 308, 313 child, 3, 6-7, 94-96, 100, 125, 260, 287,297 and fertility, 7, 20-22, 94-96, 100, 105, 109-llO, 154, 261, 328, 340, 345 fetal, 94-96, 100, 135, 337 infant, 94-96, 100, 105-106, ll3, 189, 287, 297 transition, 20-22, 105-107, 154, 250-251, 256, 263, 277, 285, 287 Mossi, 275 Mycenae, 192 Natural Fertility (see Fertility, natural) Nepal, 291 North America, 185, 218 Northwest Coast, 185 Nutrition, 108, ll5, 125, 141-142, 195, 199-200, 261, 285, 296-297, 300, 309 See also Diet, Food, Resources Paleolithic, 189 Peasants, 19-20, 196, 201-202, 207, 210, 213-214, 218, 223, 249, 255, 261-262 Philippines, 291 Pleistocene, 177-185, 189, 191, 195 Policy implications, 108-111, 226-228, 346-348 Political economy, 196, 218, 240, 248, 265, 297 Polygyny, 91, 94, 96, 100, 104, 107-108, 283, 306, 312, 337 Population decline, 217 growth, 2, 128, 196, 204, 207, 2ll, 222-223, 249-251, 279, 283, 295, 306 Postpartum abstinence (see Abstinence) Postpartum infecundability, 4-5, 7, 72-77, 105, 151, 265, 275, 283-285 See also Fertility, proximate determinants of Potato, 2, 196-197, 199-201, 207-209, 2ll, 214, 217-218 and the Irish population, 201-207 Public health, 6-7, ll5-ll9, 121, 128, 251, 263, 285 Rationality, 24, 31, 43-44, 285 arguments, 1, 30-41 criticisms of, 1, 30,34-42,55-57, 109, 156-157,257,338 empirical support for, 30, 34-42, 44-57 See also Decision Resources characteristics, 177-194 and population, 176, 291 See also Food, Income Sexual behavior, ll3, 141-142, 225, 230, 253, 275-276, 284-285, 336-337 Sex ratios, 146, 306

389 Sicily, 19-20 Singapore, 16 Slaves, 188, 254, 276, 297-300 Social class and fertility, 16, 19-20 Standard of living, 16, 194-195, 202, 204-206, 213, 216-217, 235, 248, 251, 253, 260, 313 Sterility, 107, 132-134, 141, 280, 346, 332, 335-336 Sterilization, 121, 170 Steroid metabolism, 77-79, 81-82 Stillbirths, 94, 116, 118, 135 Subjective Expected Utility Theory, 37-38 Tallensi, 275-276 Target Fertility (see Fertility, target) Technology 14 Temne, 321, 323-345, 347 Textile industry, 209-213 Thailand, 20-21 Theory of Reasoned Action, 38-41 Tikopia, 192 Tlingit, 190 Tunisia, 1, 7, 159-174 Urbanization 6 Values, 14, 126, 159-174 Value of Children Theory, 35-37 Venereal disease, 84, 86, 88, 96, 100, 108, 110 Vietnamese, 188 Wealth flows, 14 Wealth differences, 127, 129-131 Westernization 8, 13, 17, 24 See also Modernization Widows, widowhood, 64, 280, 295, 309, 334 Wolof, 317 Yukon, 186